This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
EXPLORE. DISCOVER. CREATE. o ma er what kind of camera you use, classic photography and dar roo rema·n essen ial for achieving the best possible results with today's digi al
00
s.1
boo , photographers and educators Katrin Eismann and Sean Duggan eac o use (me-tested photographic techniques with Adobe Photoshop and P Ugh room, and they illustrate each lesson with beautiful original photograp Creative Digital Darkroom offers clear, concise, insightful, and inspiring ways
0
e crea-E
photographs that express your personal vision with the highest quality s anaar s making it ideal for experienced photographers, artists, and educators alike.
Katrin Eismann is well-respected for her talent as a Photoshop expert and photographer, but above all she's considered
FEATURING: • Coverage of traditional darkroom
one of the best teachers in the field. In The Creative Digital
concepts and skills, translated into
Darkroom, she takes photographers beyond quick tips and
digital solutions using Adobe Photos hop
gratuitous effects using the clear, concise tutorial approach
C53 and Photoshop Lightroom
that made bestsellers of Photashop Masking & Compositing and Photoshop Restoration & Retouching. Katrin is the Chair of
• The most important concepts, tools,
the Masters of Professional Studies in Digital Photography at
and techniques for producing high
the School of Visual Arts in New York City.
quality images • Comprehensive chapters on black-and-white and color darkroom techniques • Clear and concise tutorials on production essentials and creative exploration • Inspiring original photography and creative insights • A companion web site offering additional information plus access to many of the book's images to help you better learn the material
Printed in Canada. Published by O'Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472. O'Reilly Media books may be purchased for educational, business, or sales promotional use. Online editions are also available for most titles (sajari.oreitty.com). [email protected].
Editor:
For more information, contact our corporate/institutional sales department: 800-998-9938 or
Beth Millett
Production Editors:
Phil Dangler, Michele Filshie, Dennis Fitzgerald
Technical Editor:
Jeff Greene
Cover Designer:
Steve Fehler
Interior Designer:
Ron Bilodeau
Print History: December 2007:
First edition.
The O'Reilly logo is a registered trademark of O'Reilly Media, Inc. "O'Reilly Digital Studio" and related trade dress are trademarks of O'Reilly Media, Inc. Adobe PhotoshopTM and Photoshop Lightroom are registered trademarks of Adobe Systems Inc. in the United States and other countries. O'Reilly Media, Inc. is independent of Adobe Systems Inc. Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks. While those designations appear in this book, and O'Reilly Media, Inc. was aware of a trademark claim, the designations have been printed in caps or initial caps. While every precaution has been taken in the preparation of this book, the publisher and author assume no responsibility for errors or omissions, or for damages resulting from the use of the information contained herein.
RepKover.. �
This book uses RepKover™, a durable and flexible lay-flat binding.
"To everyone who JeeJ the world more cLearly when looking through a viewfinde!:" -
Katrin
"To the memory of my fathe!; fameJ Duggan; friJ-hman, projectionist, hiJtory buff, camera love!; Dad_" -
Sean
Acknowledgments As Winston Churchill said, "W riting a book is an adventure. To begin with, it is a toy and an amusement; then it becomes a mistress, and then it becomes a master, and then a tyrant." Or as we have experienced, starting a book project is easy - finish i ng it is m uch more challenging; but finished it is and we can never thank the many people enough, who had the faith in us to actually get this book done.
lot more people at O' Reilly who worked on this book and without ever meeting you - we send our thanks. We' re sure you're j ust as glad to fi nally see this book in your hands as we are! Finally, behind every successful writer is an exacting development editor and we were lucky enough to work with Beth "cut that figure" Millett who paid tremendous attention to the flow and efficiency of our collaboration.
We would like to thank the photographers, both amateur and professional whose work we grew up with and that inspi red us to become photographers - from Katri n's father whose ca reful compositions with a sim ple Rolleiflex hold some of her most treasured childhood memories to Sean's father whose fasci nation with old movie cameras and stereo cameras gave Sean his first early looks th rough a viewfinder. We thank the photographers who allowed us to feature their work throughout the book; Mark Beckelman, Jed Best, Brian Nielson, Joh n McIntosh, Scott Loy, B rian Pawlowski, Michael Reichman, Jack Reznicki, and Jennifer Trenchard.
We appreciate the tremendous support that we received from the d igital imaging community. We a re indebted to the Photoshop and Lightroom teams. We admire their dedication to the continued growth of Adobe Photoshop and development of Photoshop Lightroom. Photographer Stephen Johnson found time in his busy travel schedule to review one of the final drafts and w rite the Foreword . We're honored to have his perspective in this book. Thanks also go to Dano Stein hardt from Epson; Fernando Zapata and Jim Christian from Noise Ninja; the guys at PixelGeni us for making our images j ust sharp enough; and J e ff G reene from Microsoft whose technical editing gave us two Mac-heads peace of mind.
The O' Reilly staff is excellent, including: Steve Weiss who always had creative solutions to smooth any bump in the road; Michele Filshie, who meticulously prepped the chapters before they went to layout; Ron Bilodeau whose multiple takes of the page layout created this very legible book; Dennis Fitzgerald who literally whi pped the fi nal proof into shape; and Steve Fehler who came through at the I I [h hour with a wonderful cover design. We also worked with Dawn Mann, Sarah Milstein, Betsy Waliszewski, and Colleen Wheeler - all of whom encouraged and helped us m uch more than they k now. We know that there are a
We are both fortunate to be involved in an art form that combines craft, vision, and i nspi ration. If an image is created with a plastic Holga or a high-end digital camera - we still respond to the vision of the photog� apher fi rst - the choice of technology is a distant second. As photographers and a rtists we love what we do. We sincerely enjoyed this journey and we hope you enjoy this book - may it help you to c reate the photographs in you r mind's eye.
Contents Foreword
XI
Preface
XIII
Chapter 1
Silver to Silicon
1
Seeing Images.
4
In the Digital Dark room .
6 I3
Lea rning and Forgetting the Ru les.
Chapter 2
Di ital Nuts and Bolts
15
Building a Digital Darkroom
16
Under the Hood: Essential Photoshop Preferences and Color Settings
20
Photos hop Color Setti ngs.
27
The Color Settings Dialog
29
RGB Working Spaces
29
Color Management Pol icies.
33
Conversion Options and Advanced Controls.
35
Saving Your Color Setti ngs.
36
File Navigation and Inspection .
36
Your Bags Are Packed
39
Chapter 3
Scan Develo
Iz e
41
From Analog to Digital
42
Starting with Film
43
Starting with a Digital Camera.
49
Working with Adobe Camera Raw and Photoshop Lightroom
62
Bridge and Camera Raw, or Photoshop Lightroom ?
89
tion
91 92 93
Noise Reduction Sharpening Process.
1 06
Correcting Optical Distortion and Perspective
1 09
Spotting and Cleanup.
1 17
Chapter 5
Tone and Contrast
1
Understanding Tone and Contrast .
1 30
Listen to the I mage.
1 32
Non-Destructive Editing.
1 33
Global I mage Im provements.
1 34
Converting Color to Black-a nd-White.
151
Toning and Spl it Toning Effects.
1 59
Building a Solid Foundation
171
osure Control
173 174
Non-Destructive Dodging and Burning .
1 74
Contrast and Exposure Control.
1 92
Seeing the Light
224
Chapter 7
Color Correction
viii
I Contents
227
The Landscape of Color
228
Global Color Correction
232
The Power of Curves .
246
Color Enhancements.
251
Local Color Correction
259
The Power of Lab
277
A Journey Through the Landscape of Color .
285
I
I I
t �
.
',
Chapter 8
"
�'� ' '� �
Creative Color
287
The Structure of Color ,
288
Working with Hue ,
288
Creative Color Temperature ,
30 1
Exploring Image Mode Blending,
314
Create Cross-processing Effects,
316
It's a Colorful World
319
Chapter 9
Creative Enhancements
321
Optical and Film Effects
322
Darkroom Special Effects,
332
Adding Texture,
337
Fading Away: The Distressed I mage,
339
Blending Textures into Skies
348
Creative Edge Effects,
350
Enhancing the Visual Story,
356
Chapter 10
Enhancin Focus
359
Remove Distractions ,
360
Sharpen with Finesse,
382
Valuable Noise ,
394
Your Vision Comes First,
399
Index
401
Contents I
ix
Ash Tower & Rivel; Valley of 10,000 Smokes. Katmai National Park Alasl(a. 1995; 5823 ION x J 5523 24W 697' 2560 /2:01pm; 8/29/ 95; From the project and book "With a New Eye: The Digital National Parks Project" by Stephen Johnsol1. ("I" ·1.,iJll'J'J· I. ) , 'n,,,,, '
1/, ''''
Foreword
Digital photography has now become sim ply photography. Just as desktop publ ishing became simply publishing. We may still be in a period of transition, with many improvements and critical empowerments yet to come, but photography has clearly now become an electronic medium. Software programs and computers are the tools of photographic interaction rather than the darkroom. Film and the darkroom will remain with us for many years, but they are on their way to becoming h istorical processes.
This state of affa i rs m akes it all that m uch more im portant for a book of this kind to be in your hands. Clarity and simplicity are great needs at the moment. That is where this book enters the picture. It is a workbook for today's photographer, providing needed detail to process the image and the reasoning behind the procedure. Use the pieces that you need for a particular problem or vision. Don't feel compelled by some imperative of d igital enhancement to over-work your photograph into something else.
It is an interesting time. A single company, Adobe Systems, is responsible for the most critical portal into our photographs, Adobe Photoshop, and for its up and coming sibling, Photoshop Lightroom. That has happened before in photography. Kodak was long the dominant player in the roll film market for a time, the Autochrome was the principal color process in the early years of the 20th century, and Kodachrome was practically synonymous with color photography for many years. More recently, Epson became the dominant digital photo printer company and even with Apple's fine entry into the photo editing market with Aperture, and HP and Canon's new d igital pri nters, Adobe and Epson remain the big guys. Digital photography, explained through Photoshop, has become the norm. This makes a good understanding of their d igital tools critical.
I have k nown Katrin since she was a teaching assistant for me at the Kodak Center for C reative Imaging in Camden, Maine in 1 99 1 . I saw somethi ng special i n her then, and before long she was helping to run the place. She has been speaking and writing on d igital photography for many years and has truly made a contribution of the evol ution of the medium. I may not k now Sean nearly as wel l, but his sincerity and craft comes th rough the writing and examples provided in this book.
But the current tools a re complex; many of the concepts seem brand new. The sheer number of choices can be daunting. Internet blogs and forums have people in a tailspin of convoluted procedures and in general have added to the notion of complexity. This has resulted in over-ed ited files, too much time in spent front of the computer, and too m uch distraction from the inherent value of the photograph itself.
I am glad you have this book in your hands. I'm sure it will be very helpful.
Respect for the photograph as a record of a moment before the lens is greatly needed. We've come to feel we m ust do something to the photograph in order to make it whole. Of course, this can lose sight of the only reason for the photograph to exist; the inspi ration fel t at the moment of making the image. Much can be done later to refine and craft the interpretation, but the photograph itsel f is the thing, not pixel-pushi ng later.
Katrin and Sean have put together a technical book that is a remarkably thorough and well-organized guide to the digital dark room . It is a sincere book by two good people working to make photography more accessible. Much like the software tools discussed here, in and of itself, The Creative Digital Darkroom is a good tool to keep handy.
Stephen Johnson Photographer, author Stephen Johnson on Digital Photography
Preface
As visual artists, photographers a re constantly practicing looking and seeing images. They make photographs, even when they don't have a camera with them. The eye is the lens and the mind is the camera. They notice things that others might miss, such as the way the light falls upon a child's hair, or the gentle gesture of a willow branch dipping into a stream. The small and often overlooked detail is j ust as important as the grand vista, and the unexpected afterthought often turns out to be far more significant than the carefully calculated idea. Creativity is a journey. And as we journey along that road, we become better at seeing. Seeing not only what lies ahead, but also what might lie ahead, and which visual paths are more promising than others. As we become more experienced creative travelers we become more proficient with our photographic tools and learn how to use them to their best advantage. The camera and the digital darkroom are only some of the tools we use on our creative journey. Without the most important tool, however, the creative and curious mind, even the best camera and computer system are j ust lifeless j umbles of equipment and software code. In this book we share some of the essential creative concepts and techniques that we use when working in the digital darkroom. To create a richer learning experience, many of the tutorial and exam ple i mages featured in this book are available for download from the book's companion web site, www.creativedigitaldarkoom.com (whenever you see the icon of the mouse it means that image is available for download). By working on the same images that we feature in our examples, you can practice and perfect the technique and then apply it to your own photographs. When we didn't include the exact image used, you can practice the techniques on similar images of your own. Afte r you've had a cha nce to explore our techniques with your images, we would enjoy seeing them and will post the best before/after exam ples on the book's web site. Please send them to authon@ creativedigitaldarkoom.com Please keep the attachments to under 3 MB in total.
Although the book was written for Adobe Photoshop CS3 and Adobe Photoshop Lightroom you can apply the concepts and many of the techniques with earlier versions of Adobe Photoshop, as well as Corel Paint Shop Pro or the image editing and Raw processing software you are most comfortable working with. Of course this will require that you adapt some of our instructions, but we believe you'll learn a great deal in the process. Our primary goal was to make this more than just a book of Photoshop recipes and quick tips. In addition to being photographers and writers, we a re both teachers and we want to share with our readers not only the How but also the Why. And while there is plenty of in-depth, hands-on i nstruction in digital dark room techniques covering a wide range of topics, we have also sha red our thoughts about concepts that are photographic rather than Photoshop: How to look at the image and "l isten" to it; how changes to the delicate matrix of light and shadow tones can transform the mood and meaning of a photo; how to sculpt with light and contrast; and how to imagine color and tonal changes that will best express your v ision for the photograph. One of the most rewarding and significa nt moments for any teacher is seeing the light of recognition glow in a student's eyes, knowing that a conceptual door has j ust opened, revealing new pathways and possibilities. We hope that you will experience moments like this as you journey through the pages of this book and that it will serve as a good travel com panion as you continue your photographic explorations in the creative digital darkroom . Learn from the Journey,
Please add ress comments and questions concernIng this book to the publisher:
O'Reilly Media, I nc. 1005 Gravenstein Highway North Sebastopol, CA 95472 (800) 998-9938 (in the United States or Canada) (707) 829-0515 (international or local) (707) 829-0104 (fax)
We have a web page for this book, where we list errata, examples, and any add itional i n formation. You can access this page at: http://www.oreilly.comlcatalog/9780596100476
To comment or ask technical questions about this book, send email to: authors@CJ·eativedigitaldarkoom.com
For more information about our books, conferences, Resource Centers, and the O' Reilly Network, see our web site at: http://www.oreilly.com
Colophon The book design is a result of author and designer collaboration, reader comments, our own experimentation, and feedback from d istribution channels. The text font is Granjon; the head ing font is Opti ma.
xiv
I Preface
About the Au thors
Katrin Eismann
Sean Duggan
Katrin Eismann is an i nternationally respected artist, teacher, and author specializing in creative digital photography and the impact of emerging technologies u pon professional photographers, artists, and educators. She received her BFA degree in Photographic I l l ustration from the Rochester Institute of Technology and her MFA degree in Design at the School of Visual Arts. Katrin is the author of Photoshop Restoration & Retouching (New Riders, 2005) and Photoshop Masking & Compositing (New Riders, 2006), and co-author of Real Wodd Digital Photography (Peach pit Press, 2004). She is a featured columnist i n Photoshop User magazine and is a popular speaker at the PMA, ASMP, PPA, PhotoPlusExpo, and Photoshop World conferences. I n 2005, Katrin was i nd ucted i nto the Photos hop Hall of Fame and her images have been featured in n umerous books, magazines, and group exhibits. Katrin is the co-founder and chair of the Masters of Professional Studies in D igital Photography at the School of Visual Arts i n New York City.
Sean Duggan is a fine art photographer, educator and Adobe Certified Photoshop Expert with extensive experience in both the traditional and d igital darkroom. He is the co-author of Photoshop Artistry (New R iders, 2006), Real World Digital Photography, 2nd Edition (Peachpit P ress, 2004), and he writes for national photography and imaging magazines. He has been helping photographers master Photoshop and d igital darkroom techniques for over 10 years. He teaches online classes on Photoshop for Photographers at San Francisco's Academy of Art University, and leads workshops at venues across the country such as the Santa Fe Workshops, Maine Photographic Workshops, and the University of Hawaii. He also offers customized private and online consulting sessions. H i s web site can be seen at www.seallduggall.com and he posts images as well as technical tips and m usi ngs on photography, d igital i maging a n d the creative process to his blog, wwwflpoint4.com.
Preface I
xv
Silver to Silicon
In the traditional wet darkroom, cut off f om the distractions of the outside world, i t was very common to lose tr ck o f time as we balanced working between the d ry and wet sides oft e d imly fit darkroom-e larger, coliltact heets, and yellow Kodak paper boxes on one side, and running water and scuffed developer, stop bath, and fixe," fr'tys on the other. Please don't think we are waxing poetic or that we want to be back il'l the wet darkroom for one spli t second. We certainly don't miss the fumes, the stained clothing, or the clean-up at the end of the day when aIr we wal'Lted to do was get outside for some fresh air. Yet, to this day, we both appreciate and rely on what all that time standing on the cold tile floors taught us about the craft of photography, the value of-imagination, and how to n urture a photograph as it evolved from a circled frame on a contact sheet to becoming the final print. This chapter is a summary of the tools, terms, ant! techni-Pu rge --'>- H istories. Until you get a better idea of how many history states is a good n umber for you, we suggest starting at 1 00.
Scratch Disks
When Photoshop runs out of RAM for its calculations, it grabs some em pty hard-drive space on your computer and uses that as a scratch disk. This option lets you assign a first, second, third, and fourth choice for which hard disks Photoshop should use as scratch space, as seen i n Figure 2-6. You should always assign your fastest d rive, with the largest amount of free space, to be the primary scratch disk d rive. Preferences General
Inte rface
Performance
File Handling
Available RAM: 7 3 3 MB
Performance
History States:
Ideal Range: 403-527 MB
Cursors
Let Photoshop Use:
Transparency & Gamut
�
Guides, Grid, Slices & Count
- Scratch Disks
Type
Acti...
1
�
�
I
IDrive
Cache Levels:
232.68GB
Maci. . . HD
8.27GB
H to!
( ( ( (
6K Cancel Prev Next
) ) ) )
18 - GPU S etti ngs
IFree Space "nformation
TheVault
150 16
MB (80%)
I
81
Units & Rulers Plug-Ins
- H i story & Cache
- Memory Usage
Detected Video Card:
�
ATI Technologies Inc. ATI Radeon 9200 OpenGl En gin e
� Enable 3 D Acceleration - Description Use the arrow buttons to change Photoshop's scratch disk order. Ideally, you should set your scratch disks to be i nternal drives with a large amount of free space. Changes will take effect the next time you start Photoshop.
Figure 2-6. Telling Photoshop which drives to usefor scmtch memory.
Under the Hood: Essential Photoshop Preferences and Color Setti ngs I
25
Cache Levels
Painting Cursors
Photoshop uses the image cache as a way to increase the apparent speed with which it deals with large images. Using the n umber specified by this setting, Photoshop saves several smaller versions of the image at different zoom percentages: 25, 33.3, 50, and 66.7%. When viewing the image at a zoomed-out view, such as 25%, the program can apply the c hanges to the smaller, cached 25% version first, which results in a speedier update of the screen preview. The default setting for Photoshop CS3 is six cache levels, which works j ust fine for most images from digital cameras. If you find that you're working on really large images and you have a good allocation of RAM and scratch disk space, you can try increasing it to eight cache levels, which is the maximum.
Notefor Photoshop CS Users I n versions prior to CS2, you will see a checkbox here labeled "Use Cache for Histogram in Levels." We recommend you turn this off. Although the time i t takes t o b u i l d t h e Levels histogram is shorter with this option, the reality is that you're not getting the h istogram from the ful l image, but rather you're getting a histogram rendered from whatever cached version happens to be i n use presently. We feel that i f you're going to make the effort to u nderstand what the histogram is telling you, then you should be getting the accurate data. The Histogram palette will always use the cached d ata unti l you tell it to do otherwise. It has its own interface and menu controls to render a new histogram from non-cached data.
Cursors This section of Preferences contains options that influence the display of the pixels and the appearance of the mouse cursor (Figure 2-7). Not too exciting, but there are a couple of important ones here.
From a usability perspective, this is arguably the most important setting i n this panel. By setting the painting cursors to Brush Size, you can see a circular cursor that is the size of the brush tip you' re painting with. I f the cursor is the default brush symbol icon, then you won't know how large your brush is until after you've painted on the image-very inconvenient! Sean uses the Normal B rush Tip, which shows the size of the brush out to 50% opacity (soft-edged brushes have a feathered tip that gives coverage that gradually fades out at the edges). Katrin opts for the Full Size Brush Tip that shows the coverage of the brush out to 0% opacity. Other Cursors
This preference lets you choose between standard, which is the tool icon, and a precise crosshairs. Turning this on is purely subjective, and we use standard since you can have a precise cursor tool at any time by simply pressing the Caps Lock key.
Transparency & G amut In the grand scheme of things, these preferences a re not that important, and we've rarely had to change the defaults here, but there may be times when you do need to change them, so having a basic understanding of what they do is useful. Transparency Settings
When you have a n image element on a separate layer, Photoshop uses a checkerboa rd pattern to represent the transparent pixels that surround it. You'll see the pattern only if you turn off the eye icons in the Layers palette of any underlying layers. Essentially, the program needs to have something there so you can see nothing's there. We have found that the default colors and grid size work fine for most images, but we can envision si tuations where it may be useful to change them. Clicking the colored swatches will take you to the Photoshop Color Picker, where you can choose new colors for the grid.
Preferences General
r-
Painting Cursors
Interface File
Handling
Performance
Cursors Transparency & Gamut Units & Rulers Guides, Grid, Slices & COUn! Pl ug -I ns Type
Figw'e 2-7. The Cursors preferences.
26
�
o Standard OPrecise o Normal Brush Tip o Full Size Brush Tip � Show Crosshalr in Brush Tip
I Chapter 2: D igital N uts and Bolts
- Other Cursors
�
o Standard e Precise
( ( ( (
OK
Cancel Prev Next
) ) ) )
Gamut Warning
When the Gamut Warning is activated (View ---;>Gamut Wa rning), Photoshop will place an overlay tone over any colors in the image that a re out of gamut for the cu rrent CMYK setup as specified in the Color Settings dialog. Al though its initial purpose in Photoshop was for prepress work, if you have selected an i n kjet profile as the current proofing space, it will display the out of-gamut colors for the printer and paper combination. For photographers making their own prints, this can be a very helpfu l tool for the final finessing of the image before making a print. The middle gray color at 1 00% works pretty well for color images and is not recom mended for black & white images. Click the color swatch and choose a contrasting color to cha nge it.
Units & Rulers These settings are fairly obvious (not to mention moderately yawn-inducing) so there's really no need to explain what they do. But we can offer a few useful tips that relate to choosing ruler u nits i n general and how the chosen units can affect other a reas of the program. Units
A m uch faster way to change the ruler units is to Ctrl click/right-click inside the rulers on the top and left sides of your image window. You can also change the units by clicking the small crosshair in the XY section of the r nfo palette.
The program 's measurem ent unit is used by default when you enter custom values in the Options bar for the Crop tool. Be on the lookout for this to be sure that you are not inadvertently about to crop an i mage to 8
an 8
x
x
1 0 pixels when you really needed
10 inch i mage. Simply change the "px"
abbreviation to " i n," and then apply the crop.
Column Size
If you need to resize images for a publication that uses columns for a rranging text on a page, then specifying the exact size of your columns in Preferences will enable you to resize images or create new files based on the colu m n width used in your publication. I f you want to resize a photo to two columns wide, for instance, this Column Size preference tells Photoshop how wide to make your image.
Plug - ins Most people are aware that Photoshop can be a demanding program in terms ofmemory usage (RAM). The most succinct words of wisdom we can i mpart regarding Photoshop and memory would be "the more the better ! " Apart from the amount of RAM you have, Photos hop also utilizes actual hard-disk space as a scratch disk. A scratch disk is like virtual RAM that Photoshop can use for its calculations. These preferences control how Photoshop interacts with your computer in the vital a reas of virtual memory as wel l as the location of accessory plug-ins. Plug-ins
Photos hop normally looks for filters in its own plug-ins folder (located i nside the Photoshop application folder). If you have third-party plug-ins you want to keep in a different folder, this is where you tel l Photos hop where that folder is so that the additional plug-ins are accessible to Photos hop.
P h otos h o p C o l o r S etti ngs A digital image i s nothing more t h a n a grid of numbers representing different color val ues (Sean likes to think of it as an electronic paint-by-numbers k it). U n fortunately, those n u mbers are poi ntless unless there is some standa rd defini tion of what they mean and how they should be interpreted. The Color Settings dialog is where you define what those definitions are and tel l Photoshop how it should interpret the colors in the images you bring i nto the program. We feel it serves no purpose to j ust include a screen capture of the dialog that shows you which check boxes to check and which radio buttons to enable if you don't also have an understanding of the "why behind the how" of the Color Settings dialog and of Photoshop color management in general. So, before we cover the specific options in the Color Settings dialog, let's take a moment to get an overview of the terrain we'll be exploring.
How Photoshop Approach es Color Management The core problem with having an image that is described by a grid of numbers is that it comes with a certain amount of ambiguity. To use the paint-by-numbers a nalogy mentioned earlier, a picture would look very d i fferent depending on whether you used watercolors, oils, acrylics, colored pencils, or pastels to create it. Working with d igital color includes similar variables, but rather than types of paint it is the scanner or camera, monitor, and prin ter that all describe the color a bit d i ffe rently. Even though you're working with digital val ues the actual color created, whether displayed on a monitor or printed by an inkjet or photographic pri nter, will vary from dev ice to device, because different devices interpret the numbers and render color in different ways.
Photoshop Color Setti ngs I
27
Color management i n Photoshop uses profiles to standardize how digital color is displayed and printed. Profiles simply are i n formation that is i ncluded with the file that defi nes how these colors should be i n terpreted. There are three key principles to a profile's function a lity: 1.
Having a properly calibrated display with an accurate monitor profile that tells Photoshop how your specific device renders color.
2.
Using an RGB working space that's device i ndependent-that is, its interpretation of how a given set of color numbers should be displayed is not constrained by the l imitations of a particular device, such as a monitor, printer, scanner, or camera.
3.
Adding ICC (International Color Consortium) profiles, or color tags, to your image files that tell Photoshop and other ICC-aware applications how the color n umbers in your file should be displayed. These color tags give meaning to the color n umbers in your image.
Monitor calibration and profiling
The importance of having a n accurately calibrated and profiled monitor cannot be emphasized enough. If your monitor has not been properly adj usted (calibrated) and the profile that describes it is not accurate, then no amount of color management diligence and use of image color tags farther down the line will give you predictable color. I f you have never calibrated and profiled your monitor, then we recommend postponing any serious printing of your images until you've dealt with that vital piece of the puzzle. The actual n u ts and bolts details of the calibration process are outside the scope of this book, but there are plenty of other books (Real World Color Management, 2nd edition, by B ruce Fraser, and Color Confidence, 2nd edition, by Tim Grey, are two popular and reliable choices) and resources on the Web that do cover it. At the very least, you can use software calibration to get your monitor in the ballpark, but since this relies on your subjective visual j udgment, we don't recommend the software approach. On the Mac, use the Display Calibrator Assistant, which is accessed through the Display Preferences. While not the best way to calibrate and profile a display, it is m uch better than doing nothing and it's free. As of this writing, there is still no display calibration utility built into Windows, but you can use the very capable Adobe Gamma utility that is included with Windows versions of Photoshop. We feel that the best way to calibrate your monitor (read "Do Not Pass Go") is by purchasing a third-party calibration product such as X-Rite or Gretag Macbeth, as seen in Figure 2-8. Rather than rely on the subjective and fallible calibration approach used by the Apple calibration utility or Adobe Gamma, these packages use colorimeters to measure the actual colors on the monitor, which is much more accurate.
28
I Chapter 2: D igita l N u ts and Bolts
Figure 2-8. The GretagMacbeth EyeOne Display 2 is very straightJortvmd to use and produces excellent remits on both CRTs and LCD monitors. Working spaces i n Photoshop
A working space defines how Photoshop interprets the color n umbers in any new file or in a file that you have converted from another profile. It provides visual meaning and consistency to the numbers that make up a d igital image. The working space affects any new images you create i n Photoshop a s well as images that d o not a lready have a profile associated with them (as is often the case with fi les from a d igital camera). The RGB working spaces that are available in Photoshop do not represent color as defined by a particular device, such as a monitor or printer. Because of this, they a re referred to as being device-independent. We'll d iscuss the m erits of the different working spaces a little later. As long as you're using an accurate monitor profile and savi ng an image with a color profile (more on that in the next section), then the d isplay of the image will be consistent when viewed on other calibrated and color-managed systems as well as when printed using accu rate printer profiles. (Printing issues wil l b e covered i n Chapter 1 0.) Color profiles: a d igital guide print
I f you have been photographing for a while, then at some point you probably have had to deal with photo labs to have some of your work reproduced. In some cases, when ordering an enlargement from a negative that had been previously printed, you may have brought a guide print to the lab and included it with your order to show the technician how you wanted the new enlargement to look. Without a guide print, the interpretation of a negative can be very subjective, but with a guide print, there is a reference for how the image should appear. A color profile is the digital equivalent of a guide print. It describes how the colors in an image should be displayed or printed. Along with a monitor profile and a device independent work i ng space, it represents the third
component for how P hotoshop manages the color in the images you work on. Every file should be saved with a n embedded color profile (this option c a n b e found in the Save and Save As dialog boxes). The presence of a profile, this digital guide print, tells Photoshop and other ICC savvy applications how the colors should look. We can't stress enough the importance of having a profile associated with your image. Without an embedded profile, Photoshop has no idea how to display the colors, so i t j ust interprets them according to the working space. This may represent a correct i nterpretation, but then aga i n it may not. I t's like ordering an enlargement from a negative with no guide print; the technician will simply make what she considers a good print. This hit or miss method may or may not be acceptable, while having a guide print, i.e., an accurate color profile, ensures that the image will look correct.
The Co l o r S etti n gs Dialog Now that we have some of the important background information covered, let's take a look at the actual Color Settings dialog and discuss some of the options there. You can find the Color Settings near the bottom of the Edit menu (Cmd/Ctrl-Shift-K), as seen i n Figure 2-9. Color Settings
� '\SP
Synchronized: Your Creative Suite applications are synchronized using the same color settings for consistent (olor management.
- OK
{ Nonh America General Purpose 2 ' :1 r- Working Spaces :::-;::;;; : ;;;:;:;::;;;:::;::;:= : =====;;-] RGB : ( sRGB I[C6 1 966 2.1 :I CMYK: ( U.S. Web Coated (SWOP) v2 :I Gray: ( Oot Gain 20% !:I Spot: { Oot Gain 20% :I Settings:
-
Cancel Load... Save ...
( More Options )
� Preview
r- Color Management Policies
( Preserve Embedded Profiles [ :1 CMYK: { Preserve Embedded Profiles f: I Gray: I Preserve Embedded Profiles t : I Profile Mismatches: D Ask When Opening O Ask When Pasting Missing Profiles: D Ask When Opening RG8:
r- Description
North America General Purpose 2: General-purpose color settings for screen and print in North America. Profile warnings are disabled.
Figure 2-9. The Color Settings dialog box looks more daunting them it really is.
At the top of the Color Settings dialog box is the Settings menu, which contains some preset configurations tailored for different purposes. I n Photoshop CS2 and CS3, the default settings are North America General Purpose 2 . These default settings don't portend t h e e n d of the world in terms of color quality, but if you do care about tonal and color quality in your images, they're not ideal. As we will explain shortly, the primary downside to these defaults is the use of sRGB as the RGB working space.
As a shortcut to get to most of the settings that we recommend for photographic work, you can choose North America Prepress 2 from the Settings menu. In the fol lowing section, we'll address what these settings mean, explain why we think you should use them, and cover situations where using a d i fferent setting may make sense.
RG B Wo rki n g S paces For t h e purposes of this discussion, w e a re going t o focus on two types of RGB worki ng spaces that you can use i n Photoshop. The fi rst of these are what w e wil l refer t o a s the standard working spaces. These are the four working spaces that are available in the Color Settings dialog when it is in Fewer Options mode. The other working spaces are k nown as wide-gamut spaces because they encompass a much larger gamut than the standard spaces. These become available only when you press the More Options button. Although a larger gamu t may sound like the way to go (after all, more is better, righ t ? ) , larger gamut working spaces come with their own set of challenges and caveats. We will start off with the standard spaces and discuss the wide-gam u t alternatives in a bit.
Standard RG B Working Spaces I n the default a rrangement of this dialog box (Fewer Options), you h ave four choices for RGB Working Spaces in the pop-up menu. Of these, only two a re serious contenders for photographers who care about good color reproduction. Unfortunately, neither of them is used in the default settings. Let's take a look at these choices in greater detail in the order in which they appear. Adobe RGB (1998)
Of the four standard RGB working spaces, this is the one often recommended for most photographers. (If you chose North America P repress 2 from the Settings menu, Adobe RGB is a l ready selected for you.) This working space has the largest color gamut of any of the fou r standard spaces. For printed output of your photographs on RGB devices, w hether on ink jet printers or on photographic printers such as a LightJet 5000, Adobe RGB encompasses a good portion of the color gamuts of those devices, and we feel i t i s the best choice of the standaJd vanilla RGB working spaces in Photoshop. Please see "Wide-Gamut Working Spaces" on the fol lowing page for additional recommendations.
Apple RGB
This is a legacy working space. I n Photoshop 4 and earlier, the only color space the program used was based on an Apple 1 3-inch monitor, hence the inclusion of Apple RG B in the list of choices here. There is no reason to choose this as a working space.
RGB Worki ng Spaces I
29
ColorMatch RGB
This working space is based on the gamut of an actual device, the Radius P ressView monitor that was once ubiquitous in prepress shops. Although the gamut of Color Match is much smaller than that of Adobe RGB, it does include most of the common CMYK gamuts, and it can be a logical choice in certain situations, especia lly if you're preparing images specifically for press reproduction. We also know some portrait photographers who prefer this working space because skin tones do not appear as saturated as they do in Adobe RGB. sRGB
This is Photoshop's default RGB worki ng space, so if you've never changed it, or if you just accepted the defau l ts when you installed the program, i t's probably still set to this. Microsoft and Hewlett-Packard developed the sRGB space to represent the gamut of the "typical" monitor. Since the "typical" monitor is probably an inexpensive one that is not designed for imaging work, sRGB is less than ideal for people who are concerned about working with color photographs. For the casual photo hobbyist, sRGB is fine, but if you have purchased this book, your i n terest i n photography and the digital darkroom is likely more serious than that of a casual hobbyist. We do not recommend sRGB as a working space.
sRGB clips substantial colors in the blue/green range that can be reproduced on even matte surface inkjet prints (matte papers have a smaller gamut than glossy). To be fair, Adobe RGB also clips some colors that are within the gamut of some matte papers, but the clipping is minimal compared to sRGB.
Wide-G amut Working Spaces: ProP hoto RG B I f you press the More Options button, you will see a list of every RGB profile installed on your computer. Although seeing so many potential choices may be a daunting prospect, the good news is that nearly all of them a re entirely inappropriate as an RGB working space because they are device profiles (that is, profiles that represent a specific monitor, scanner, or printer).
30
I Chapter 2 : D igita l N uts and Bolts
There is one RGB working space in this list, however, that you may want to consider as an option, especially if you are shooting RAW files and a re i n terested in retaining all of the color and tonal range that can be captured by you r camera. ProPhoto RGB is a wide-ga m u t space that was originally specified by Eastman Kodak as a way to describe all of the highly saturated colors that could be produced by E6 transparency films. It is much larger than Adobe RGB and encompasses nearly all of the colors i n the visible spectrum (as well as some colors beyond it). Before you rush off and start using it, however, you need to be aware of the special handling that it requires. Although it is a very powerful color space, to l i ft a q uote from Spiderman's Uncle Ben, "with great power comes great responsibility." Advantages of Pro Photo RGB for raw files
The reason ProPhoto RGB is of i n terest i f you a re shooting raw is that it encompasses nearly all of the color information that your camera's image sensor can capture. Even though Adobe RGB is the largest of the fou r sta ndard working spaces in Photoshop, converting a raw file into that space will clip significant amou n ts of color information that the camera is capable of capturing i n some scenes, especially in the deep, saturated colors. This does not mea n tha t you can't get a great pri n t from a ra w image that has been processed into Adobe RGB, but it does raise this question: Why th row away all of this i n formation i f you don't have to, as illustrated i n Figure 2 - 1 0 ? Several years down the road, we may have output options that will allow for la rger color gamuts. Even today, the newer Epson inkjet printers and K3 inks are capable of printing satu rated cyans, magentas, and yellows that lie outside the gamut of Adobe RGB. By intentionally clipping the file at the beginning, as it is converted to Adobe RGB for your master file, you are closing the door on those possibilities. You can al ways return to the original raw captu re and reprocess it, of course, but any work done to the layered master file will have been al ready forced i n to a much smaller color gamut. If the image is converted into the ProPhoto RGB color space at the beginning of the process, however, virtually all of the original color and tonal range will be preserved in the master file (Figure 2 - 1 l ). This makes it ideal to use as an archive color space so that your master images are stored with as much potential i n formation as possible. Working with your raw files in ProPhoto RGB enables you to "future-proof' your images.
Using ProPhoto RGB with 8-bit files significantly increases the risks of posterization or banding, especially when major edits a re performed . To get the most out of what ProPhoto has to offer, you need to be converting your raw images to 1 6-bit files (don't even bother using it with J PEGs). We should poi nt out that we do not consider 'vvorking in 1 6bit to be a downside at all (in fact, we prefer it! ) , but some people may, and you do need to be aware of this as it relates to this color space.
Figure 2-10. The grid represents the Adobe RGB worl(ing space, and the color shape represent>· the color profile of the Ep,ion 2400 when printing to Luster pape!: Notice how the essential green>' and yellows filll outside of the working space,
Figure 2-1 1, In ProPhoto RGB all of the pom'ble color>' of the EjHon 2400 printing to Luster paper are maintained, ProPhoto RGB disadvantages: Here there be dragons
When j udged by the basic observation that ProPhoto RGB preserves a much fuller range of the captured i n formation whereas Adobe RGB clips significant amounts, it may seem like a no-brainer to use ProPhoto RG B all the time. Although this is certainly a compel l i ng reason to consider this color space, there are good reasons to proceed with caution. As a working space ProPhoto RGB is not for everyone or for every image. While it does offer more possibilities for preserving all the color and tonal information of a raw capture, when used i ncorrectly, there is also the potential for mistakes that can make an image look very bad. Let's take a quick look at some of the possible downsides and other issues to be aware of.
ProPhoto RGB encompasses most of the visible spectrum, but it also exceeds it in some areas of deep bl ues and deep greens. (Some have referred to these as "imaginary" or "science fiction colors" because they do not really exist.) When editing an image, it is possible that colors can be shifted into these fictional ranges, which cannot be seen or reproduced , and this can cause major problems in the colors you can see. Because wide-gamut spaces are so wide, they are far larger than what can be displayed on a monitor. This being the case, some changes you make to an image in the ProPhoto RGB, especially s h i fts in hue or satu ration, may not be able to be distinguished on a display. The debate about whether ProPhoto RGB makes sense in real-world situations (as opposed to the purely theoretical) continues to rage on in the photographic com m u n i ty. Many photographers question the relevance of using a color space that is vastly larger than your eventual output space. While this poin t has some merit, the following is also true and worth consideration: Some camera and "scene" gamuts can contain colors that fal l outside of Adobe RGB, as seen in Figure 2- 1 2 . (This is especial l y true with highly saturated yellows and greens that are often found in flowers and foliage on a bright, sunny day.) If you convert a raw file to Adobe RGB as you bring it i nto Photoshop, the original character of those colors is lost when they are com pressed in to the smaller gamut Adobe RGB space. With the newer Epson K3 inks and printers, some colors that can be reproduced on those printers fal l outside of the Adobe RG B space. And it is not improbable to think that future printers and inks will be capable of la rger color gam uts than what is possible with today's models. If your goal is to capture and reproduce as many colors as faithfully as you can, then for some images, ProPhoto RG B may be a better choice than Adobe RGB. ProPhoto RGB does not clip those colors at the begi n n i ng d u ring the raw conversion process, and you can exercise greater control over how the 1 6-bit ProPhoto image is converted into the pri nter space at the end of the process.
RGB Worki ng Spaces I
31
the individual pri nting press. For these reasons, i t's not as easy to give a one-size-fits-all recommendation for the best CMYK space to use. For general purposes, the default U.S. Web Coated (SWOP)v2 is probably as safe a choice as any for a default setting. Since CMYK settings are so tied to how the job will be reproduced, however, any setting you choose here should be thought of as no more than a placeholder that wil l suffice for the most common printing situations. Depending on the work you do and the type of publications in which your work appears, one of the supplied presets may be j ust fine, but you should always maintain good channels of communication with your publisher or printer and verify that you are using the right setting.
Figure 2-12. A rawfilefrom a Canon 20d, as mapped into Adobe RGB worl(ing color space, reveals clipping in the greens and some yellows.
We do not recommend ProPhoto RGB for everyone. I f you d o not have a good understanding of working i n 1 6bit, color management fundamentals, profile conversions, rendering intents, and soft-proofing, this is probably not the color space you should be using at this time, and Adobe RGB would be a safer choice. But if you a re interested in preserving as much of the original color captured by the camera as possible and have the requisite Photoshop experience to n avigate the potential pitfalls, then ProPhoto RGB may be worth considering for your high-bit images that originate from camera raw files.
CMYK Working Spaces You need to be concerned with choosing a CMYK worki ng space only if the images you're working on will be reproduced on a commercial printing press and you will be applying edits and color corrections to the actual CMYK files. If you are working with RGB files only, then you don't need to trouble yourself with this setti ng and you can leave it at the default. Some people think they need to use CMYK for their desktop inkjet printers but this is not the case. Even though ink jet pri n ters do use various combinations of cyan, magenta, yellow, and black inks, they a re considered to be RGB devices because they do such a great job of converting the RGB i n formation you send them. While RGB working spaces are device-independent, CMYK is rigidly output-specific. The flavor of CMYK that you use is i n fl uenced by the type of inks and paper being used for a given project and, in some cases, the cha racte ristics of
32
I Chapter 2: D igita l N uts and Bolts
We believe that color i mages should be kept in a n RGB working space as long as possible, and flattened copies converted to CMYK only for spec i fic purposes (that is, it makes no sense to have a "generic" CMYK file that you provide to your stock agency because it may not be suitable for all types of CMYK reproduction). If you fin d that you need to delve i n to this a rea fu rther, we highly recommend Real World Adobe Photoshop CS2 by David Blatner and Bruce Fraser (Peach pit Press, 2005) as a n excellent resource for anyone who needs to use Photoshop with CMYK.
Gray Working Space G ray working spaces can be selected to reflect specific dot gain characteristics or to d isplay gammas. Dot gain percentages refer to the fact that when printed on a press, a dot of ink wil l increase in size, and therefore become da rker, as it is imprinted and absorbed i n to the paper. The gamma settings are designed for images that wil l be viewed on a monitor, but they also work well for images that will be printed on an inkjet printer. If you r primary output is to a printing press, then choose a dot gain that matches the same figure in your CMYK setup. A setting of 20 percent is a common percentage for coated paper stock, for instance, and is the default in Photoshop CS3. As with CMYK, however, dot gain may vary depending on the pa rticular inks and paper being used, so consult your printer to get as much information as possible. If you're printing black-and-white images to a desktop ink jet printer, we recommend setting the G ray working space to Gray Gamma 2.2. This is true even if you're on a Mac, which still uses a default d isplay gamma of 1 .8. A gamma of 2.2 (here and in your monitor calibration) will produce smoother gradients than a gamma of 1 .8, and it more closely matches the tone response curve of most d isplays.
If you're on a Mac a n d open an earlier grayscale
file created using a gray gamma of 1 .8, you will be notified that the embedded profile does not match the current Gray working space of 2.2. In that case, j ust choose "Convert document's colors to the working space," and the tones in the photo will be converted with an eye toward preserving the appearance of the image.
Spot Working Space Spot Working Space refers to very specialized prepress situations where custom inks (also k nown as spot colors) or varnishes will be used. This is even more of a prepress concern than basic CMYK and u nless you fi nd yourself having to deal with custom inks and varnishes (perish the thought! ), the default setting of 20% is fine. If you need to use it at all, then do check with your printer about the specific dot-gain characteristics of the ink, paper, and press that will be used to print the j ob.
Color Manage m e nt Po l i c i es This section of the Color Settings dialog tells Photoshop how to behave when it encou nters images that don't have a profile (untagged) or images that have an embedded profile that does not matc h the currently selected working space (mismatched). This is the place that controls those a nnoying messages that sometimes a ppear when you open an image in Photoshop. Well, some people feel they're annoying, but once you understand what they're tel ling you, they're not so bad. There a re three separate pop-up menus for setting the policies for RGB, CMYK, and Gray working spaces. All contain the same three choices: P reserve Embedded Profiles, Convert to Working RGB (or CMYK, or Gray), and Off. These choices are the same ones that appear in the warning dialogs when you open images (though the exact wording is a bit different). What you select here j ust determines which radio button is selected by default when the warning dialog pops up (you can still change it i f you want to). Let's take a close look at exactly what these choices mea n. We'll look at them in order of appearance: Off
Before we get into the nitty gritty about what this choice does, let us state up front that choosing this is a bad idea (cue foreboding and scary m usic). I f you open a file that contains an embedded profile, Photoshop will discard the profile and regard the image as u n tagged. The color
numbers in the file wil l be interpreted according to the currently selected working space, even though that may not be a correct assumption. Remember the analogy we used earlier about a profile being the digital equivalent of a guide print? Selecting "Off' here would be like the photo lab th rowing away your guide print after you had placed the order! Since we believe that profiles (if they are accurate) are a good thing that can hel p you control the color in your images, it's probably no surprise that we don't recommend this option . By stripping the profile from the image, you a re flying blind as to the true meaning of the color n umbers, and Photoshop can display the file only according to the specifications of your working space. Apart from obscure color management testing purposes, there is no good reason to choose this setting. Preserve Embedded Profiles
When opening a file with a color tag that is different from your working space, this option will honor the embedded profile, and no conversion will be performed on the color numbers in the file. The image opens in Photoshop, and you can work on it in its own color space without having to convert to your working space. Assuming that you have a properly calibrated and profiled monitor, the display of the image should be correct. This setting is usefu l i f you get files from different sources and want to make a conversion to your working space only after you've had a chance to inspect the file. Convert to Working RGB
When you open a file that has a color tag different from your working space, this choice will convert the image from the embedded profile into your cu rrently selected RGB working space with the goal of preserving the appearance of colors in the image. If you' re working in a closed loop system (in other words, you're generating all of your images with your camera) and you k now where all of the files are coming from, you know they have accu rate profiles, and you're using a working space such as Adobe RGB or P roPhotoRGB, then this is probably the best choice for most photographers. Of the above options, we recommend Convert to Working RGB (or Working G ray) as the best choice in most circumstances. Remember that all these settings do is determine w hich button is preselected for you in the profile messages that appear when you open a fi le; you can still make a different choice if it is appropriate for a specific image. For the CMYK working space, it p robably makes more sense to choose Preserve Embedded Profiles.
Color Management Pol icies I
33
Profile Mismatches and Missing Profil es
Opening Files: How to Deal with Profile Warnings
These th ree checkboxes control whether or not you see those vexing "missing profile" or "mismatched profile" notices when you open a fi le (Figure 2 - 1 3). If you never want to be bothered by them again (and we hear this sentiment a lot! ), you can turn them off here. We recommend that you leave at least two of them on, however, since we think it's always good to be informed about w hat's happening with the color in our file or how it's being interpreted. The most important are the Ask When Opening options that trigger a notice if you open a file that either has no profile or that has a profile that doesn't match your working space. We feel you can safely turn off the Ask When Pasti ng option. This triggers a notice when you paste from one image to another and the profiles of the two don't match. In nearly all cases, you will want to convert the color numbers so the image appearance is preserved. Since that is the default choice for this warning, we feel it's fine to leave this option unchecked and let Photoshop do the conversion without bugging you about it.
One of the most common questions we get from students and n ew Photoshop users is what to do about the missing profile and profile mismatch warnings that often pop up when you open a file. I f you don't k now what they' re tell i ng you, or what the right answer should be, encountering these can be very frustrati ng. To make matters more perplexing, they use language that is slightly different from that used in the Color Managemen t Pol icies section of the Color Settings. In an effort to clear up some of the confusion surrou nding these warnings, here a re some recommendations on what choices are appropriate when you run into them. Profile Mismatch
Profile mismatch warnings are the easiest to deal with because Photoshop detects that there is a profile associated with the image and this gives it the necessary information to convert the color numbers from the image's existing profile to match your current RGB working space. Of the th ree choices presented to you, only the first two are really an option, as seen in Figure 2 - 1 4:
Color Settings
�
E
Unsynchronized: Your Creative Suite applications are not synchronized for consistent (olor.
Settings:
(
r:l
ProPhoto for I6-bit RAW
- Working Spaces RGB: CMYK:
-
I ProPholo RGB
( ( (Dol
U.S. Web Coaled (SWOP) v2
Gray: Gray Gamma 2,2 Spo!:
Gain 20%
- Color Management PoliCies
:1 :1 :1 :1
OK
( ( ( ( �
Cancel load . . . Save ...
Fewer Oplions Preview
(
� Ask When Opening � Ask When Pasting 51 Ask When Opening
Missing Proflles:
I
Engine: Adobe (ACE)
�
l
Intent: Relative Colorimetric
Advanced Conlrols
r- Description
:1 :1
�% 11.00 I
ProPhoto for i6-bit RAW: This configuration is used for Raw captures that are processed as l6-bit files. Uses ProPhoto RGB as the RGB Working Space and Gray Gamma 2,2 as the Gray Working Space. Preserves all embedded profiles as the default,
I Chapter 2: D igital N uts and Bolts
Working: Adobe RCB (199B)
Cancel
) (�_O=-K,--..J)
Figure 2-14. A profile mismatch is a common occurrence that should not induce cold sweats or uncontrollable panic.
I
Figure 2-13. Checking Profile Mismatches and Missing Profiles informs you when a profile conflicts with your color settings.
34
Embedded: ColorMatch RCB
(
� Use Black Point Compensation G!1 Use Dither -ize la,-ge enough to accommodate the 1Cl1-gest size print you plan to mal(e. Then that master image can be repUiposedfor many different projects_ �JSI)
48
I Chapter 3 : Scan, Develop, and Organ i ze
Working with reasonably larger sizes gives you the leeway to make larger prints. When working with an art di rector or designer, working on files that are up to 25% la rger than requested allows for some cropping or enlarging i n the page layout phase. Using a larger file to make smaller files will always look better than taking a small file and sizing it up to make a big print. Of course this doesn't mean that every file you work with should be big enough for a billboard, but you should consider making the master file larger than you think you'll need, as you'll never regret the extra i n formation.
Starting with a Di g ital Cam e ra The megapixel resol ution is the main feature bantered about when a camera's technical specifications a re mentioned. This number refers to how many m illion s of pixels can be recorded by the image sensor, and there is a d i rect correlation between the sensor's resolution and the size of the file it produces. But resolution and the quantity of pixels are only two aspects of digital capture that affect i mage quality. Additional variables that impact image q uality i ncludes lens q uality, ISO, camera fi le format, lighti ng, shutter speed, whether a tripod was used to steady the camera, and atmospheric conditions. Knowing the others and how to control them is an important fi rst step i n ensuring that you a re starting out with the best image possible.
Dig ital Camera Resolution Determining a digital camera's resolution is a simpler undertaking than calculating scan n i ng resolution and file size because there is a specific limit to how many pi xels your camera can capture. The i mage sensor is a specific size measured in X and Y coordinates, from the tiny sensors i n your camera phone t o t h e multi-megabyte resolution o f the professional medium and large format scanning cameras. But j ust as fil m resolution is i n fl uenced by photographic variability so is d igital camera performance. In fact, you could use a pinhole body cap on a high-end d igital SLR and produce a file with less image i n formation than the simplest disposable camera would deliver. We respectfully partake in a moratorium on the debate that compares film to d igital solely on the grounds of resolution. There is more to a photograph than resol ving line pairs or rendering test targets or bragging about file size. Photography is an art form and a means of creative expression and we should use and explore all available tools and techniques to express our aesthetic. One of our mottos has always been to learn the rules, learn them well, and then break them ruthlessly as Figure 3-9 illustrates. Sean photographed a scene and then re-photographed the image with a variety of technologies to create a soft, low detail image that is q uietly melodic.
Figure 3-9. This image was originally shot with a 12-megapixel digital SLR. The file fi"0112 that camera was then displayed 011 a laptop screen and rephotographed with the same digital camera through the domed glass viewfinder ofa Kodal( Duaflex twin lens reflex camera datingfi"0112 the 1950s. « '.\/1 H ow many pixels do you need?
The most useful strategy we can recommend is to always shoot at your camera's maximum image size since that gives you the most flexibility i n terms of what you can do with the image. This is especially true if you a re i nterested in maki ng large prints, or i f you need to crop and show only a portion of the frame. The more pixels you can capture, the more options you will have. We advocate shooting in raw for most images, and raw files are always captu red at maximum size. But even i f ci rcumstances dictate that you shoot i n JPEG (i.e., fast-moving sports, pa rty pictu res or quick snapshots), you should always choose the largest resolution and h ighest quality setting (least amount of com pression) available. You probably parted with a lot of money to buy your d igital SLR, so i t makes no sense to use only a portion of the capabilities you paid for by shooting at a smaller image size. Use all of the viewfi nder
Since the n umber of pixels you can record is d i rectly related to how la rge a print you can make and still expect good q uality in terms of detail resolution, i t ma kes the most sense to compose and frame your i m ages using all of the area in the viewfi nder. Any time you shoot wide and crop later, you a re essentially l owering the camera's resol ution (Figure 3- 1 0). Whenever the situation allows, fill the viewfinder with the composition you want to achieve, to take advantage of all the p i xels your camera has to offer.
Starting with a D igital Camera I
49
I t's not that you can not get good images when shooting JPEG, but your exposures need to be much more exact since J PEG does not have the tonal flexibi lity of a raw file. For those with experience shooting film, J PEG is like slide film, while raw is more akin to color negative fi l m . With slide film, you have to be more exacting about your exposures while negative film offers greater exposure latitude and much more flexibility for interpreting the negative and making a print. And, like a color fi l m negative, a raw file can be rei n terpreted many d i fferent ways. Finally, only by shooting in raw can you gain access to the high-bit data that your camera's image sensor can capture. At the time of this writing, most high-end digital SLRs that shoot raw can capture a 1 2-bit file, whereas the J PEG format can support only 8 bits of data. Figure 3-10. Rather than rely 012 cropping in the compute,; capture as many pixels as possible by using as much ofthe viewfinder as possible for your compositions. For the close-up view of the dool; Sean tool( a second photo andfilled the viewfinder with the wine/ow and the bel/, rather than rely on cropping the wider view. If'lS/J
Fil e Formats Th ree primary file formats are available to users of digital SLRs and even some compact cameras: J PEG, TIFF, and raw. The main d i fference between these formats is that both J PEG and T I FF a re "finished" images, where the camera has processed the file before writing the file to the memory card. With J PEG, there is also additional file compression that disca rds color data to reduce the file size written onto the camera card. A raw file is not a fin ished image; it is the recipe to make a n image and you have a tremendous amount of control over how to interpret the recipe when you convert the image using Adobe Camera Raw, Photoshop Lightroom, or other raw conversion software. It is this ability to interpret the image with such a high degree of control that makes raw our preferred file format. Depending on the type of photography you do, there may be situations where the l ighting is controlled and never cha nges, such as high-volume catalog photography, in which JPEG makes more sense, or i n which shooting in raw would be overkill for the demands of the job and what the client requires. But i f you want to be sure that you a re capturing the best set of image data possible, then using raw is the best approach.
50
I Chapter 3 : Scan, Develop, and Organ i ze
The value of h ig h - b it files
The highest possible number of discrete tonal levels that a standard 8-bit file can conta i n is 256 per color channel. Adj usting the tonal range of a d igital image involves some loss of data as you move the existing i n formation a round the tonal scale to get the i mage looking the way you want. For images that requ i re critical, h igh-quality tonal and color correction, 256 tonal levels per channel do not offer you a lot of editing room to enhance tone and color without degrad ing the tonal i n formation. This is especially true on images that have less than opti mal exposure and require what Sean refers to as "heroic" color correction (see the sideba r "The Vegas Synd rome: Why Bit Depth Matters"). Working with h igh-bit data-files that contain more than 256 shades of gray per chan nel (see table 3.2)-gives you more tonal values to change the tonal and color character of an image. Mainta i n i ng your files i n a high-bit state as long as possible ensu res visually h igher q uality prints, especially in the delicate highlight transitions and subtle shadow a reas. Table 3-2. Bit-Depth per Channel
Tonal Levels
8-bit
256
l a-bit
1 ,024
1 2-bit
4,096
1 4-bit
1 6,384
1 6-bit
65,536
The Vegas Syndrome: Why Bit Depth Matters An 8-bit file can contain no more than 256 separate tonal levels per color channel. A 1 2-bit raw file can contain up to 4,096 tonal levels. This is significant because all pixel based, digital tonal adj ustments i n Photoshop i n volve loss of tonal i n formation as you shift the values around to correct the expos u re.
Think of i t this way: Let's say you're playing blackjack i n Las Vegas and you have $256 to gamble with. After a while, you have lost $ 1 00. That represents a large chunk of your entire gambling budget. I f you come to the table with $4,096 in gambling fun money, however, then the loss of $ 1 00 is not such a big deal. You still have plenty of dollars (tonal values) left to play with (Figure 3-1 1 ).
lI:. I H istogram before adjustments (8- and 1 6-bit)
1 6-bit H istogram after adjustments FigU1·e 3-1 1. The same set ofimage adjustments (Levels and Curves) was applied to an 8-bit and a 1 6-bit version ofthis image (original on the left; adjusted version on the right). The top histogram is how the tonal 'r ange looked before the adjustments. The middle one is the 8-bit file after the adjustments and the bottom one is the 16-bitfile after the adjustments. The gaps in the 8-bit histogram indicate a loss oftonal data, which could show up as posterizatioll in the shadow areas.
A digital camera that shoots i n raw can capture high-bit data ( l 2-bits), and we recommend you take advantage of the additional tonal i n formation that h igh-bit data provides. Most good scanners also enable you to access h igh-bit data. As the old saying goes, i t's better to have i t and not need it, than to need it and not have it. At the very least, try to perform all tonal and color corrections and d ust-busting on the high-bit file before converting a copy (save the master 1 6-bit version with a d j ustment layers i ntact) to 8-bit for further editi ng. With a raw file, as many global corrections as possible should take place in the raw conversion process. From a tonal quality perspective, the best workflow would be to keep the file in h igh-bit for the enti re digital dark room process.
Working in high-bit does require larger hard d rives and a healthy amount of RAM, however, because I 6-bit files are twice the size of8-bit files and the addition of a lot oflayers can cause the file size to expand rapidly. Additionally, not all the filters work on high-bit files, though the most important ones for photographic work a re supported. With our own images, we try to work with the high-bit information as long as the file size and processing time is manageable. If we reach a point where we no longer need to have a high-bit file (such as when all tonal correction is finished and we're ready to start compositing data from multiple images), then we choose T mage -;.Duplicate and check Duplicate Merged Layers Only to flatten the duplicate file. We then convert to 8-bit with Image-;.Mode-;.8-bit and make sure to save this new file with a new name.
Starti ng with a D i gital Camera I
51
Exposure and Histograms Resolution and file formats are tech nical settings that you choose i n the camera. Capturing a good exposure and correctly understanding what the camera's histogram is telling you is more closely aligned with your skill as a photographer. And j ust because Photoshop exists to help you finesse your i mages to perfection, we strongly believe that it's best to try to get it right in the camera first and not rely on "fixing it" in Photos hop. The less you have to do to an i mage in Photoshop, the better. Your work in the digital darkroom will go faster and smoother, leaving you more time to be out creating photographs with your camera, which is much better than sitting i n front of your computer! This is not a book about digital photography, but since this section of the chapter is about i mage quality at the beginning of the process, it's worth a l ittle page space to review a few important concepts. The histogram
The h istogram display on a digital camera is a powerful tool for analyzing exposure i mmediately after you have taken a photo, enabli ng you to make changes to either aperture or shutter speed that will result in a better exposure. It is j ust a simple bar graph that shows how the tonal values are distributed across the tonal range (Figure 3- 1 2). On the far left is a total black with no detail (level 0) and on the far right is a total white with no detail (level 255). The height of the bars indicates the number of pixels present in that part of the tonal range.
I
Black o
Midtone 1 28
I
White 255
Figure 3-12. The histogram.
The h istogram is the tonal signature or fingerprint for an image. There is no u niversal perfect h istogram shape that you should strive for, since each image is d i fferent and contains its own specific arrangement of shadow, mid tone ,and h ighlight values (Figure 3- l 3). I n a very dark image, such as one taken at n ight, the h istogram would naturally have much of the data grouped on the far left side, j ust as a bright, high-key image would show more of the tonal information on the right side.
52
I Chapter 3 : Scan, Develop, and Organize
Figure 3-1 3. There is no ideal histogram shape that is appropriate for all images. Each histogram reflects the tonal signature ofthe photograph.
I n working with the h istogram to rev iew and a nalyze exposure, there a re three main a reas to be concerned with: preventi ng h ighl ight or shadow clippi ng; ensuring that you have a good exposure that captures a wide range of tones, especially on the right side; and if you a re shooting in raw, for some types of photography there a re additional exposure considerations for capturing the best possible set of tonal i n formation. Clipping control. [n terms of exposure, the important thing
is to make sure that there is no clipping (tonal values that have been recorded as either a total black or a total white) in areas of the image that contain significant detail. [f clipping does appear in an image, it shows up in the histogram by the vertical bars running right up against the left or right sides, and ending abruptly, almost as if they had been cut off (Figure 3- 1 4). Clipping is usually more of a problem in the highlight areas simply because they are lighter tones and are immediately visible in a print. However, if vital detail is present in the dark shadows that you want to preserve, you need to keep an eye on the left side of the histogram as well. If you see clipping when you review the histogram on the camera, ad just the exposure to give the shot either less or more exposure as the situation demands. This can be accomplished in several ways, including using exposure compensation if you are shooting in a semi automatic mode, manually adjusting aperture or shutter speed, or even metering and exposing the scene in a different way. Controlling clipping, especially i n the highlights, can often be a challenge in high-contrast scenes. The current collection of digital SLRs has a dynamic range that is closer to slide film than negative film. I f you have ever shot slide film, you k now that brightly lit, h igh-contrast scenes can be d i fficult to expose, and often a comp romise m ust be made w here the exposure is biased towards controlling the highlights while sacrificing some of the shadow details. This is another reason to shoot in raw, as you have many more options
d u ring the conversion p rocess for c reating a n i mage where the h ighl ights are not blown out but you still have good detail i n the dark shadows. Even though both Camera Raw and Photoshop Lightroom do have the ability to recover some detail in blown-out h ighlights, there a re li mits. Get your exposures correct in the camera and you'll have better results. Captu ring good tonal detail. By following good exposure practice in the camera, you m i n imize the chance that your images will have problems that need further corrections. This is especially relevant i n terms of i mages that are too dark. It is m uch easier to darken a photo that seems too light than it is to lighten an underexposed, dark i mage. B r ighteni ng up the dark a reas of an underexposed scene can introduce excessive gapping i nto the h istogram, which can manifest itsel f in the image as posterization or banding, w here you don't have a smooth transition between tonal levels. And when a n underexposed i mage is lightened, there i s also a very good chance that you will reveal noise i n the shadows. While Photoshop and other software have noise reduction solutions that may be able to m i n i mi ze the problem, the results are never as good as you get with a n image that had no noise to begin with. The best strategy i s to avoid as m uch noise as possible in the first place and one way to do this is to shoot properly exposed i mages. This may not be possible in all shooting situations, of course, especially those where you find yourself in "available darkness," but i t's a worthy ideal to aim for. Please see C hapter 4 for n umerous techniques in avoiding, h id ing, and reducing noise. Exposure considerations for raw files. D igital sensors record data in a l i near fashion. This means that the brightest "stop" of exposure, which is represented by the data on the far right side of the h istogram, contains fully half of the available tonal levels. In a typical camera with a dynamic
Figure 3-14. In the image on the left, shadow detail in the darl(est areas of the rocl(S is clipped. In the image on the right, the highlights in the railing and parts ofthe sky are clipped. «J SI)
Starti ng with a D igital Camera I
53
range of approximately six stops, this translates to 2 ,048 val ues, half of 4,096. In basic aperture theory, each time you stop down the lens to the next smallest aperture, you let in half as much l ight. This applies to how a sensor records data, too. Since each stop of exposure below the brightest represents half as m uch tonal i n formation, the next stop after 2 ,048 tones would contain 1 ,024 values and the next stop down from there is halved to 5 1 2 . The stop afte r that contains 256 values, then 1 28, and finally 64 levels in the da rkest shadows. You may have heard that it is a good idea to underexpose your photos so you do not risk the highlights being blown out to total white. While this practice will control the highlights, it also means that you a re not taking advantage of all the available tonal levels that your sensor is capable of recording. I f the histogram data does not extend into the far right-hand side, you are using far fewer tonal levels than the camera is capable of capturing. While a n u nderexposed raw file can certainly be lightened in the raw conversion process, this means that the 64 to 1 2 8 levels that represent the da rker shadows need to be stretched out and redistributed into the brighter areas of the tonal range and on some exposures this may result in noise and posterization appearing in the midtones or shadows. You want to have the data in the histogram as far to the right side as possible without any clipping (Figure 3- 1 5). Using this method, the image may appear too bright and washed out on your camera's LCD screen, but this can easily be adj usted in the raw conversion process without any of the side effects that come from trying to lighten up the dark shadows on an underexposed shot. This method of exposure is well suited for landscape and fine a rt photography. Si nce it relies on recording scenes with more exposure, however, and more exposure is achieved either by openi ng up the lens or using a slower shutter speed, it may not be a feasible practice for photographic situations where you need to capture fast
action, or where good depth of field is requi red and a tripod is not practical. Digital exposure is all about capturing the best set of information about the scene. The expose-to-the-right method is one way to structure the exposure to capture the fullest range of tones possible, but it often requi res more post processing in the raw conversion step. Whether you choose to use the expose-to-the-right method or not, it's important to u nderstand that a good d igital exposure is similar to a film exposure in that the goal is to place the tonal values for very l ight and very dark subjects so that there is good rendition of those details. With a good set of tonal i n formation recorded you have more latitude and flexibility in processing that information as you work towards the goal of creating a file that the print media can handle. Exposure considerations for J PEG fi les. 1 f you are shooting JPEG, your goal should be to expose the i mage so that it looks the way you want it at the time of captu re or is slightly overexposed without any h ighlight clipping. It is better to bias the exposure more towards the l ight side because, as explai ned in the previous section, d igital sensors record more tonal values in the h ighlight range than in the shadow range. I f a n image is a bit too bright (but the h ighl ights are not clipped), it is not a problem to d arken i t down. But brightening an i mage with significant u n derexposure problems can reveal even more problems.
Using Color Reference Targets A fair amount of color correction practice is built on the foundation of identifying neutral tonal val ues. Simply put, i f you can locate a tone in the image that should be a neutral (that is, no color cast at all-a gray, white, or black) and then apply adj ustments that ensure that the tone is neutral, m uch of the color cast problems i n a n image can be quickly solved. One way to speed up this process is to shoot the fi rst images of a setup with a color reference target in the frame.
Figure 3-15. The exposure for the image on the right is biased more towards the important right-hand side of the histogram, which meal1J" it has used more of the j'emor's available capacity for recolding tonal data.
54
I Chapter 3 : Scan, Develop, and Organ ize
Color references are available at any good professional photography resource and usually consist of a card that contains a grid of precise color and gray patches. The most wel l-known color targets a re the Gretag Macbeth ColorChecker, WhiBal®, Kodak Q- 1 3, and ExpoDisk. By referring to the neutral and color patches on the chart as you evaluate the image that contains the chart, you can see how well color is being reproduced in the image. The gray patches are especially useful si nce you can use them to quickly establish a good white balance setting for an entire session of photos. Fi rst, you adjust one image that includes a color or gray reference that was photographed in the same lighting and same setup as the rest of the shoot or in a series of shots. Then you can save that group of settings or a pa rticular subset of settings. In Adobe Bridge or in Adobe Photoshop Lightroom, you then select all of the other images that were shot in the same conditions and batch apply the settings from the color reference image (Figure 3- 1 6). Although it may seem that color reference ta rgets a re suited more for stud io-type photography, they work well for any type of photographic situation where you have the time to
take a couple of test i m ages that contai n the reference card at the begi nning of a series and a re especially useful when you a re working with m i xed light sources.
F i l e O rgan izati o n a n d Arch ive Pre parati o n I f you have been shooting with a digital camera for even a short amount of time, you k now how easy it i s to amass a h uge quantity of i m ages. And with the mega p i xel count on the march u pwards, this also translates to larger file si zes and more storage req u i rements. Before we get to the c reative image e nhancement part of the book, we feel it necessary to spend some time on the essential file management part of the process. While this may not seem as exciting as trans forming an i mage with careful use of adj ustment layers a n d some well -placed layer masks, it is an essential part of the workflow i f you want to easily fi nd your images again and also if you want to ensure that they a re thoughtfu l l y p repa red for a n archive that is well orga n i zed and as fut u re-proof as possible.
Figure 3-1 6. Using a color reference target can help in determining a white balal/ce setting that can be quiC/(ly applied to a group of images that were all photographed in the same lighting. 1 / )
F i l e Organ i zation and Arch ive Preparation I
55
There are several steps in this process and whether you decide to do them in sequence all in one session or parcel them out in stages over several days or weeks depends on the number of images you have to process as well as how m uch time you have to devote to the project. In this section, we'll take a look at some of the key tasks related to importing images and initial file management and organizational matters. •
I m porting
•
Batch Renaming
•
Making Selects with Ratings and Labels
•
Applying Keywords
•
I ni tial Adj ustments to Select Raw Files
•
Converting to DNG
As an important part of this section, we'll d iscuss broad concepts rather than the details of specific software programs. Although we will be referring to current programs such as Adobe Bridge and Adobe Photoshop Lightroom to illustrate how you can choose to manage the fi rst phase of working with your images, it is the nature of software that it does change and evolve, especially in this developing field of d igital photography workflow. Any fi le management protocols you decide to adopt should be flexible in that the basic tasks a re not specific to any one software program.
able to create the thumbnails and previews, you know the files are OK. The memory card can then be reformatted in the camera with the camera software.
Batch Renaming The file names assigned by the camera are not that useful. After the photos have been downloaded to your computer (and backed up to at least one other location on a separate drive), the fi rst thing to do is to rename them with a more meani ngful name. Many programs, including Adobe Bridge, have batch renaming features where you can select an entire folder of images and apply a new name with variables such as custom text, date created, and sequential numbers in a variety of configurations. Some applications, such as Adobe Photoshop Lightroom, let you rename files as they are downloaded from the memory card. If you a re using such a program in your file management workflow, renaming the files may be handled at the initial import stage (Figure 3- 1 7). Importing 1 7 images [hilt were all taken on May 20, 2007. File Handling:
Copy To: /Users/sfd/Piaures
Organize:
Importing This is the stage when your images are the most fragile, because they exist only in one place-the memory card. When importing your images, you should create a d uplicate set of all the files on a separate hard d rive as soon as possible. Some photo utility programs, such as PhotoMechanic or I magelngester, will i m port images from your memory card and automatically copy them onto other d r ives. Photoshop Lightroom also offers this functionality. In addition to their l aptops, Sean and Katrin both travel with compact Firewire and USB hard d rives for creating m ultiple i mage backups. The laptop and hardd rives are never kept i n the same bag, in case one bag is lost or stolen. Digital "wallets" that allow you to download d i rectly from a memory card without a computer are also quite useful for this purpose. I f a separate hard d rive for backup is not available, then consider burning CDs or DVDs as temporary backup. The important thing is to make sure that right from the start you have a minimum of two copies of your downloaded images, preferably on separate hard d rives. You should not reformat the memory card until you have verified the integrity of the imported fi les. An easy way to do this is to have Adobe Bridge or Adobe Photoshop Lightroom build thumbnails and previews of the files. Since these are generated from the actual files and not from camera thumbnails, if B ridge or Adobe Photoshop Lightroom is
56
I Chapter 3 : Scan, Develop, and Organize
(
�
,
{ By Original Folders
lS �
)
Backup TO:
/: \ ... ( Choose ) ': 1
! Copy Photos as Digital Negative (ONG) and Import
070520_Pmhale-TinyTeaParty
-
Ignore Suspected Duplicates
VOlumes/SD Image Arc. .ghtroom-Import Backups
17
-
C Choose )
File Naming: Duggao_070520 9405.dng Template:
� Duggan File Name Templatel
1
f:
)1]
Figure 3-1 7. In Adobe Photoshop Lightroom, you can renamefiles as they are downloadedfrom the memory card and create a bacl(up copy . to a separate hard drive.
The ini tial inclination may be to rename the files using a content-based description that tells what the photo is of, where it was taken, or what the name of the client is. Specifying content in a file name can get cumbersome at times, however, and adding keywords to the file's metadata is really a more flexible way to attach content descriptions and other terms that can help you find the images quickly. Sean uses a naming system suggested by Peter Krogh in his excellent title, The DAM Book: Digital Asset Management for Photographers (O'Reilly, 2005). His fi le naming convention is his last name, the date the image was created, and the original camera identifier number. So an image shot on August 24, 2007, would be named like this: "Duggan_070824_3825. CR2" (the date is listed in the YYMMDD format).
Making S elects with Ratings and Labels After renaming the files, the next step is to sort th rough them and apply some form of notation to distinguish the great shots from the near great, the good ones, the merely so-so ones, and the ones that are obviously destined for the trash. Doing this early in the process is a very val uable investment of your time that pays off down the road when you need to review groups of images and n arrow your search by targeting only those images with specific labels or ratings (Figure 3- 1 8). By restricting the image search to only those that have th ree or more stars, for example, you immed iately have a smaller group of images to rev iew that you have al ready ranked as good shots. Adobe B ridge and Adobe Photoshop Lightroom both use the color label and a star system with the ability to apply up to 5 stars to an image. When using color labels, you may for exam ple designate the red label as Move to Trash. While sorting through your images, this red label is applied with a si ngle keystroke. Later you can choose to view only the images with a Move to Trash label to review them one final time before actually deleting them.
as in cataloging applications such as Microsoft Expression Media (formerly iView MediaPro) and Extensis Portfolio. The addition of keywords can hel p you locate images very quickly, enabl i ng you to perform searches based on keywords rather than looking for a specific image by file name or date created . For i nstance, you may have several images of covered bridges but a re looking for one special shot. I f you ha ve adopted a keyword i ng system you could do a search for all images tagged with the keyword "covered bridge" and your i mage catalog program would display only the images that had that keyword. With all of your covered bridge images displayed, you could q uickly locate the special shot that you had in m i n d . Searching for the keyword "covered bridge" in an image catalog program quickly displays all the image files in the archive that have been tagged with that keyword (Figure 3 - 1 9).
B y Filename By Document Type By Date created By Date file modified By File size By Dimensions By Resolution By Color profile Label
Figure 3-19. Keyword searches quid(lyfind tagged images. Figure 3-18. Once ratings have been applied to a group ofimages, it is easy to view only a subset ofthose shots that have a higher rating. Above: The Filter pane in Adobe Bridge.
When applyi ng star ratings, you should use the higher star ratings sparingly. Such a rating system is only effective if you are giving h igh-star ratings to only the very best images. And not all images need to have a star rati ng. I m ages with no rati ngs applied essentially have a " neutral" value - they are neither good enough to have a positive star rating nor are they bad enough to be labeled as an outtake or a reject.
Applying Keywords Labels and ratings a re the fi rst level of metadata information you can add to your images to help you organize them. The application of keywords is yet a nother, more detailed level that lets you add to your files specific i n formation that can be used in Photoshop Lightroom, Adobe Bridge, as well
Keywords can be applied i n Adobe Photoshop Lightroom , Adobe B ridge, and other programs either individually or, more efficien tly, to many images at once. Depending on the program you use, you may also be able to apply keywords when the i mages are downloaded and imported i nto your digital workflow program (Figure 3-20). Like the appl ication of rati ngs and labels, adding keywords during the initial image orga n i zation process does take a little extra time, but it is a worthwhile time i nvestment that pays off very quickly once you start searching your i mage catalog based on the keywords you have added. The add ition of keywords to your files gives you a great deal of control over your image collection that makes it easier to ma nage and increases its value. Add i tionally, you get more enjoyment from your image archive once you can locate files without spending hours searching for that one perfect shot.
F i l e Organization and Arch ive Preparation I
57
r. r.
Import Photos
Importing 1 5 4 images that were all taken on Apr 18, 2 007. :
File Handling: ( Copy Photos as Digital Negative (oNG) and Import
Hawaii, Pearl Harbor. USS Arizona, memorial Check All
Render Standard-Sized Previews
( Uncheck All )
(
� Show Preview
Cancel
) (
Import
)
Figure 3-20. Applying general keywon/s as a group of images is imported into Adobe Photoshop Lightroom. Tn this case, the four keywords being used am apply to all ofthe photos being imported. Late}; more specific keywords can be added as needed.
For personal work, you should be able to find a n image with a few basic keywords. Katrin uses a general to specific a pproach: Place - Location - Subject - Attribute - Emotion. For example for the image seen in Figure 3-2 1 - the place is Germany, the location is the Black Forest and the name of the town (spelled in two d i fferent ways) and the subject is crucifix. r n this case, she uses Lightroom to easily find all images of crucifixes as seen in Figure 3-22, which also includes two shots taken in San Francisco. For professional image stock work, each stock agency requires a very specific keywording method that can easily require 20 to 30 keywords per file.
For
a
and
file
thorough naming
discussion strategies,
controlledvocabulary.com .
of check
keywording out
www.
Figure 3-2 1. The general-to-specific I{eywords for this image are Germany, Blad( Forest, Sa!zJtetten, Salttown, and crucifix. '"
58
I Chapter 3 : Scan, Develop, and Organ ize
adjustments. Using this approach, you can adj ust one file that is representative of the e ntire group, and then, i n Adobe Photoshop Lightroom or Adobe B ridge, apply those settings to other files that can benefit from the same adj ustments (Figure 3-23). We'll cover making adj ustments to raw files in both Camera Raw and Adobe Photoshop Lightroom a li ttle later in this chapter.
Figure 3-22. Using Lightroom to peli01111 a .,-earch for all images tagged with "crucifix".
Initial Adjustments to Select Raw Files After labels, ratings, and keywords have been applied, the next step is to make i nitial adj ustments to select raw files. This is when your work applying star ratings starts to pay off since you should finesse only the i mages with higher star ratings. This approach saves you a lot of time since you can ignore most of the images from a specific shoot and concentrate on only the very best ones. Adj ustments to raw files are global i n natu re, meaning they apply to the entire image. Your aim here is to quickly add ress global issues such as setting highlight and shadow points, overall brightness, contrast, and color saturation. For specific local corrections, such as l ightening or darkening particular a reas, you have to wait u n til you bring the image into Photoshop, where selections, adj ustment layers, and layer masks can be used. Where possible, you should minimize your work by seeing how many i mages can benefit from the same set of
o Treatment (Color) Color Saturation Vibrance
BlaCK Clipping
o
Brightness
Split Toning
Color Adju.stmenu
Spot Removal
o Str.1ighten Angle
Crop
o
Aspect Ratio
Contrast lens Corrections Chromatic Aberration
Ton e Curvt
lens Vignetting
o Sharpening
o Luminancf!
Noise Reduction
o
Calibration
o Color
( Check All) (Check None)
� (Synchronize)
Figure 3-23. In Adobe Photoshop Lightroom, the adjustment settings for thefirst sailboat image are synchronized to the other sailboat Images.
Proprieta ry Con version Programs More and more camera man u facturers are using thei r own software conversion solu tions as a way to sell cameras. A good example of this practice is Nikon's Capture NX, which lets the user make local a reas adjustments to raw files. J ust as with regular global adj ustments that can be made i n many raw conversion programs, these local adjustments are not permanent and can be changed at any time. The one caveat, however, is that you can use the software only on Nikon raw files. While the
program may deliver on everything that it promises, and from a marketing standpoint it may be an effective strategy, we are a l ittle leery of the proprietary nature of such applications, especially i n terms of creating and maintaining image a rchives that will be accessible i n the future (more on that i n the next section). For that reason, our focus in this book is on software that can be used on files from a wide variety of cameras, and not j ust those from a single manufacturer.
File Organ i zation and Arch ive Preparation I
59
Converting to D N G One o f the most importa n t steps i n our import and organ ization workflow is to convert our camera-specific raw files to the d igital negative format (DNG). Depending on your workflow, this can be done at the begin n i ng of the process or at the end. The DNG format is of special interest to d igital photographers because it is an openly documented raw format that has d istinct advantages over the proprietary raw format that your camera creates. Before we get i nto how to convert fi les to the D NG format, let's take a brief look at j ust what the DNG format is, why it was created and what it offers you. Can yo u r raw files be opened i n the future?
Every camera that can shoot in raw uses i ts own file format that is specific to that camera model. Every time a new camera comes out, a new raw file format is introduced. These formats are proprietary and not openly documented, which means that the camera makers do not divulge all the details about how the files are written or any special capabilities that are engineered i n to them (some raw formats have features that can be accessed only by software created by the camera manufacturer). I f the camera makers do not share the details of the file format, then companies that make raw conversion software have to get hold of a camera, c reate test i mages, and reverse engineer the fi les to figure out the best way to decode the files. The concern among many photographers, curators and historians is that the undocumented nature of raw files does not guara n tee that those files can be accessed in the future. If the camera vendor were still around, one would hope that they still support fi les from older cameras, but there is no guara n tee that they will do so, or that third party companies will always write software that supports older raw formats. Software that converts raw fi les may also be tied to specific operati ng systems. D N G : an open source raw format
In 2004 Adobe Systems created the DNG format specifically to deal with concerns about the future accessibility of raw fi les. It is a n openly documented raw format and the i n formation requ i red to program a DNG converter can be obtained by anyone at no cost. This reason alone makes DNG very attractive as a n archive format. It is more l ikely that you will be able to open a n openly documented raw format many years i n to the future than the undocumented, proprietary formats created by camera vendors. Advantages of D N G
T h e fact t h a t a DNG r a w file is openly documented and more likely to be accessible for many years i n to the future is the primary advantage, but there are others.
60
I Chapter 3 : Scan, Develop, and Organ i ze
No X M P fi les. Camera-created raw files requirea separate XMP (Extensible Metadata Platform) file to hold added metadata and adjustments applied in Camera Raw. This separate sidecar file is not required for DNG because the format was written specifically to be able to include such information, without ever permanently altering the actual raw data. Additionally, keywords and other metadata can be written back into a DNG file after it has been created, even from other applications, such as asset management programs like Expression Media. Ability to embed the original raw file and previews. If
you are unsure of totally getting rid of your original camera vendor raw format (that is, the .NEF format for Nikon or the .CR2 format for Canon) you can choose to embed a copy of the original camera vendor raw file in the converted DNG file. This original raw file can then be extracted later i f you needed it. The only reason to do this would be if at some point you wanted access to vendor-specific metadata tags or functionality that the vendor's raw conversion software offered, or prove that the original camera file had not been altered. The one potential weighty downside to this is that it does increase the file size, but the option is there for those who want it. Unless you work in an evidence based field such as forensics, police or museum reprod uction and have to prove that the file is the original camera file, you probably don't need to embed the original raw file into the DNG file. You can also choose to embed a full-sized J PE G preview that can be used by some asset management programs to quickly generate proofs, web galleries, or display your adjusted raw file accurately in a photo-cataloging program. Those programs do not understand the XMP files generated by Camera Raw so any adjustments you make in Camera Raw, or other programs such as Adobe Photoshop Lightroom or Apple Aperture, a re not reflected w hen they create thumbnails and preview i mages. If you want to use other programs to quickly generate proofs or content such as web galleries or presentations from the DNG file, then we suggest saving a medium-size J PEG p review when the file is converted to DNG. If you will be generating 8 x 10 or larger sized prints from the embedded JPEG i mage, then the full size preview would be the better choice. Disadvantages of D N G
While t h e D N G format h a s m a n y advantages, there a re some thi ngs to be aware of. The main one is that at the present time, the major camera manufactu rers do not support it. This means that once a Nikon NEF file has been converted to DNG, you cannot use Ni kon software to adjust it (if the original raw file was embedded into the DNG, then this can be extracted in order to work with i t using camera maker software). While a few camera manufacturers do currently offer DNG as a n i n-camera raw format, i t reall y needs to
be adopted by the major players-Nikon and Canon-as a raw format i n the camera. The scene is a bit d i fferent for raw processing and photo workflow software, however, a nd at the time of this writing nearly 50 programs offered varying levels of DNG s upport.
Exporting XMP Data to DNG Files in Ligh troom 1 . 2 a n d Earlier
ONG Conversion in Photoshop Lightroom
In Lightroom, there are two methods for converting your raw files to DNG. To convert at the time the photos are imported i n to the Lightroom libra ry, open the File Handl i ng menu in the Import Photos dialog and choose "Copy Photos as Digital Negative (DNG) and I m port"(Figure 3-24). This gets the DNG conversion out of the way at the beginning of the process and is one less thing to do later on. Also, si nce DNG files are approximately 30% smaller than raw files, you'll save some hard disk space as well. The one disadvantage to converting to DNG as the photos are imported is that it takes longer than sim ply i mporting the raw files with no conversion. I f you're under time pressure to quickly review the shots from a photo shoot then converting at the time of i m port may take too long. To convert a group of images to DNG after they have been added to the Lightroom libra ry, choose Libra ry-"Convert Photos to DNG (Figure 3-25). Since the DNG format is designed as a raw format, it is best to leave the option to "Only Convert Raw Files" selected. I f you want Lightroom to delete the original raw files after successful conversion to DNG, there is an option for that. 7 4 photos . taken from J a n I S . 2007 t o J a n 20. 2 0 0 7 file Handling:
:
Copy Photos as Digital Negative lONG) and Import
:
1
Copy To: /Users/Sean/Pictures Organize:
I
:1
Into One Folder
fY1 Put in Subfolder:
Bone Shop
Figure 3-24. Copying and wl1vertingfiles to DNG upon import into Lightroom.
� Only convert RAW files o Delete originals after successful conversion Cancel
) (
There are a few under-the-hood settings and menu options regarding XMP metadata to be aware of. The first is in the Metadata section of the Catalog Settings (File-"Catalog Settings). It controls whether changes are automaticall y written i nto XMP (Figure 3-26). We recommend that you turn this off since Lightroom has to w ri te every change you make at the time you make it and with DNG files this can really slow the program down. With this preference u nchecked, however, you need to be s u re that you Export the XMP data i nto the raw or D G fi les once you a re done working with them (ideally at a time when you a re not using Lightroom to process images and the program can update the files without the slowdown impacting your work). To do this, choose Metadata-" XMP-"Export XMP Metadata to Files. 600
Catalo g Settings
{ General File Handling Metadata I , Ed� -,� 9-=-----:----:---:-�-__;:::;===:;:::===�--, ' tjn": 1 � Offer suggestions from recently entered values ( Clear All Suggestion LIsts )
(0
g Write develop settings to XMP for JPG, TIFF, and PSD
Convert Photo to DNG
(
Any changes you m a ke to image files in Lightroom (i.e., image adj ustments, keywords, etc.) a re stored as a set of i nstructions i n the Lightroom database. I n order for these changes to be v isible in other applications, such as asset management programs l i ke Extensis Portfolio, those changes need to be written back i nto the file. For raw files, the changes are a utomatically written i n to a n XMP sidecar file. For DNG files they are written d i rectly i nto the DNG "wrapper."
OK
)
Automatically write changes into XMP V/arning: Changes made in Ughrroom wH/not automatically be visible in other applications.
Metadat� Br(101ner Categories � Camera � lens � File Type
� Aperture � Shutter Speed
IS1 1s0
� Date � location
� Creator � Label
1
J
Speed Rallng
Figure 3-25. Converting to DNG after images have been added to the Lightroom library. Figure 3-26. To avoid worliflow slowdowns with DNG files, unchecl{ the option for automatically writing changes to XMP. Just be sure to expol·t the XMP data into the files once you {/1'e done wm'king with them.
File Organ ization and Arch ive Preparation I
61
. The D N G Converter
Although you can use Lightroom and also the Save dialog i n Camera Raw t o convert files t o the D N G format, for batch processing la rge n umbers of files, the free DNG Converter avail able at www.adobe.comlproductsldngl is much more convenient (Figure 3-27). You can specify an entire folder of files and place them i n to a separate desti nation folder without overw riting the original raw files. Plus, as a stand-alone application, the DNG Converter can work on the conversions while you a re free to do other work in Photoshop, Bridge, or Camera Raw. eeo
c: o
ONe Converter
Adobe' Digital Negative Converter
Select the images to convert ,. Select Folder... /Users/sfdlOesktopLOAM Workflo...60930_SeaTruk-PortClyrle-Maine/
) ( [Y1 lnclude images contained within subfolders
e
Select location to save converted images
o
Select name for converted images
( Save in New location �; I ( Select Folder... ) /VolumnJDeriv31ives_Ol/_lmageWmConvl!:rt!.d to ONG-ToArchive-BU/ 1!1 Preserve 5ubfolders
Name example: MyDocument.dng
!Document Name
Begin numbering: File extension:
o
Preferences
------;1 �
ill + I 0+
( Change Preferences...
JPEG Preview: Full Size Compressed (lossless) Preurve Raw Image
•
•
•
I nitial raw adj ustments, quickly applied as part of the import and sorting process Batch adj ustments to many i m ages from the same sessIOn Adj ustments to single images as a precursor to further work in Photoshop
Several raw conversion applications are on the m arket. Some a re stand-alone, raw processors, and others work i n conj unction with other products o r as a part of a broade r digital photography workflow solution. I n this section we'll take a look at working with raw files in Adobe Camera Raw and Adobe Photoshop Lightroom.
Adob e Camera Raw Camera Raw is a part of the Photoshop CS3 and Adobe Bridge software package. It works with either program, which means that you can work with your files in B ridge without having to launch Photoshop. Access the Camera Raw controls by opening a raw file from a s upported digital camera. The easiest way to do this is to target a folder of raw images in B ridge. Double-click any raw file and Camera Raw i nterface will appear with a preview of your image.
Camera Raw is periodically updated to add support for new cameras that were released after the
Don't embed original
( About DNG Converter ... ) (
and med itative interaction with the i mage and can occur at different poi nts in the workflow:
Extract...
)
Quit
)(
Convert
F(gure 3-27. Thefree Adobe DNG Converta
shipping version of Photoshop CS3. You should always check the Adobe Web site to ensure you are using the most recent version. This is especially important if you purchase a new camera that came
Wo rki n g with Ad o b e Cam e ra Raw a n d P h otosh o p Lightro o m The great flexibility to i n terpret files i n differen t ways, create derivative versions of the same image, or salvage a less than optimal exposure is what makes the raw format so compelling for photographers. We are very fortunate to live at this poin t i n photographic h istory, as this is the type of control over the image that photographers could only d ream about before now. The actual work of processing and i n terpreting the raw file can be quickly applied or i t can be a more considered
62
I Chapter 3 : Scan, Devel op, and Organ ize
out after the release of Photoshop CS3. Depending on how new your camera is and when the last update was posted, you may have to wait until the next Camera Raw update before you can use ACR with raw files from your new camera.
Camera Raw interface
When Camera Raw fi rst opens you may see a yellow triangle with an exclamation poin t in the upper-right corner of the preview image for a few seconds. This is j ust an i n dicator that Camera Raw is still building an accurate preview from the i n formation in the raw file. The image may appear a bit u nsharp and pixilated while the preview is generating but it will look sharper when the preview is complete (Figure 3-28).
White Balance and Color Sampler tools Zoom and Hand tool
Crop and Straighten tools
Retouching tools Preferences dialog
H igh light clipping display
Full Screen Toggle
Shadow c l i p p i ng display
Exposure I nfo Control Tabs
4850
Settings Menu
�
c AltlO
:)
!1.mYll C
-0. 15 0
Image Settings c C
+50 +25 0
Saturation
C C
+1 +,
, Open Image ) � � Zoom Level
I mage Preview Area
Workflow Options
Figure 3-28. The Adobe Camera Raw 4. 1 dialog box. Camera Raw tools, preview and clipping displays. J ust above the preview image on the left side is a series of tool icons. The first two control basic navigational tasks such as zoom ing and scrolling. The eyed roppers a re used for setting a white balance and placing color sampler poi nts to track important tonal values. The Crop tool lets you apply a non destructive user-specified crop, or choose from a l ist or preset aspect ratios. The Straighten tool is used for straightening crooked images. The retouch ing and redeye tools a re new to Camera Raw 4.0. The former lets you apply non-destructive and reversible d ust spotting to your images. Even better, you can synchronize your raw settings and apply them to a group of similar images of the same scene, enabling you to effectively handle d ust spot removal as a batch operation. The last buttons in the tool bar will open the Camera Raw preferences and rotate the preview image.
toggles back and forth between the adj usted and origi nal settings of the current control tab panel. You can also toggle the Preview by pressing P. To see an overall before and after preview open the Settings menu on the u pper right corner of the control tab panels and switch between Camera Raw Defaults and Custom Settings The small triangles in the upper left and right corners of the histogram turn on the Shadow and Highlight cli pping display. This will show you when shadow tones a re forced to black (blue overlay color) or h ighlights that a re totally white with no detail (red overlay color). You can also activate the cli ppi ng d isplays by pressing U for shadows and 0 for highl ights. These alert colors will be visible only i f you're viewing an image where this loss of detail, or cli ppi ng, is present, or i f your adjustments a re forcing shadow or highl ight values to be clipped (Figure 3-29).
Controls for Preview and switching the dialog box to full screen view a re to the left of the histogram. The Preview
Worki ng with Adobe Camera Raw and Photoshop Lightroom I
63
0
Contrast
0
Clarity
0
Vibrance
0
Saturation
0
Figure 3-29. The overall clipping di,-play colors ,-how,- red where highlight values are clipped and blue where shadow values are ciljJped.
r0-
O &u2
LlillI!!1
� 0 0
�O.75
0
rsillJ +25 0
Saturation
0 ::
0
OI�
Figure 3-30. Highlight clipping can be diJ'jJtayed on a per-channel basi,- by holding down the Option/A It I(ey and clicl(ing the Expo,-ure slider. A/'eas that turn white indicate clipping in all three color channels.
64
I Chapter 3 : Scan, Develop, and Organize
Temperature ---0 Tint
0 &!\.Q
Exposure
� 0
Recovery C Fill light 0
=m Brightn
2 Si
(
ComraS[
0.00
0 0
+50 +25
Clarity O�=-==�=-�-=��-= Vibrance
Saturation
o o
Figul'e 3-31. Shadow clipping can be displayed on a per-channel basis by holding down the Option/Alt I(ey and clicking the Blacks slidel: Areas that turn black indicate clipping in all three color channels.
Though the cl ipping warning can be useful and it is certainly important to k now when this type of clipping is happening, the one d rawback of this d isplay is that it turns on the overlay color even i f only a single channel is clipped, al though there may be useable and recoverable detail in the other two channels. We prefer to show clipping by holding down the Option/Alt key while clicking and d ragging the
Exposure or the Blacks sliders to th reshold the i m age that uses a m ultiple color d isplay to show you the clipping in each individ ual color channel or i n channel pairs (see the sidebar "Decoding the Clipping Display Colors"). Figure 330 shows a n example of h ighlight clipping while Figure 3-3 1 shows an exam pIe of shadow clipping.
Decoding the Clipping Display Colors When the clipping d isplay is activated by pressing Option!Alt and d ragging the Exposure slider left or the Blacks slider right, the colors i n dicate where clipping is occu rring i n the photo and i n which channels. Here's the key to u nlock the color code: Highlight Clipping:
Black means there is no clipping. White indicates clipping i n all three color channels. Red, green and blue indicate clipping in each of those channels. Yellow shows clipping in the red and green channels. Magenta shows clipping i n the red and blue channels and cyan shows clipping i n the green and blue channels.
Shadow Clipping:
White shows areas with no shadow clipping. Black i n dicates clipping in all three color channels. Cyan shows clipping in the red channel, magenta i n the green and yellow ill the blue channel (cyan, magenta and yellow are the opposite of red, green and bl ue and i n dicate the total absence of those colors). Red, green or blue shows shadow clipping in the other two channels. For instance, red shows dipped shadows i n t h e green and blue channels; green indicates cl ipping in the red and blue channels, and blue reveals dipping i n the red and green channels.
Working with Adobe Camera Raw and Photoshop Lightroom I
65
The clipping display in Photoshop
I mage Adjustments a n d Workflow Options. Below the histogram and the Settings menu are the eight tabs of the I mage Adj ustments (Figure 3-33). These a re where you make the changes to your raw file. From left to right they are:
Lightroom
works the same way. Holding down Option/Alt and clicking the Exposure slider will show clipping for the highlights and doing the same for the Blacks slider will show shadow clipping.
•
The h istogram and settings m e n us. On the right side of
the dialog is the h istogram, which shows you how the tones in the file are distributed along the tonal range. Below the histogram on the left are the RGB val ues for the cu rsor location as you move it over the preview. Next to the RGB values is the exposure i n formation for the image. On the right side of the d ialog box at the top of the ad j ustment tabs is a small button for accessing the Settings menu (Figure 3-32). This contains options to use I m age Settings, which are settings that a re associated with the cu rrent image if that file has al ready been adj usted i n Camera Raw, or revert to the Camera Raw defa ul ts for the camera that created the image. This latter choice will remove any adj ustments you may have made to a file. Custom refers to any new settings you have made in the current work session with an i mage. You can also apply the Previous Conversion if those adj ustments will work for the current image. The lower part of this menu contains additional choices for work ing with Camera Raw settings.
lillfAl ,,,, ���Jlkll'.Jl� Basic
White Balance:
tAs Shot
Temperature
Exposure
Recove.ry
FiJI Light Blacks Brightness Contrast
"46'5'()
�
Tint
&lli!
Default
ib ,
r ;=� �
� ,�
l
age Sertings
Camera Raw Defau Its Previous Conversion
Custom Sertlngs
Preset Sertings Apply Preset
0:00
rorors--
--:;:so
l+2S
Export Sertlngs to XMP
•
•
•
•
•
•
Tone Curve, for precise ton a l and contrast adj ustments using Cu rves Detail, which enables you to apply sha rpen ing and noise red uction HSL/Grasycale, for adj ustments to the h ue, satu ration and lumi nance of colors, as well as custom grayscale converSIOns Split Toni ng, adapts a classic dark room technique to the d igital age and al lows you to apply a d i fferent color tone to the highl ights and the shadows. Lens, for correcting lens-based errors such as chromatic aberration and vignetting Camera Calibration, which enables you to create custom profiles for how your camera behaves in certain l ighting conditions. Presets is where you can create and save your own Camera Raw settings that can be quickly appl ied to Images.
Under the preview image at the bottom of the dialog box is an underlined summary of the current Workflow Options. Clicking this will open the Workflow Options dialog box (Figure 3-34), where you set the Color Space that the image is processed i nto (for example, Adobe RGB and ProPhoto RGB), the bit depth of the file, the pixel dimensions of the file and the resol ution. We'll d iscuss the I mage Adjustments and the Workflow Options i n more detail i n the next section.
Update DNG Previews load Sertings... Save Sertlngs... Save New Camera Raw Defaults Reset Camera Raw Defaults
Figure 3-32. The Camera Raw settings menu.
66
�
Clear Imported Settings
•
Basic, which controls white balance, exposure (the white poi nt), recovery (for rescuing overexposed highlights), fill light, blacks (the black poi nt), brightness, contrast, clarity, vibrance and saturation, and fill light
I Chapter 3 : Scan, Develop, and Organize
The Auto adjustme n ts
I n the Basic tab, two underlined options for Auto and Default appear above the Exposure slider (Figure 3-35). The one that is gray and unavailable as a clickable choice indicates the setting that is currently being used fonhe image. The Defaul t settings are the base line Camera Raw interpretation for images from each camera, representing an i nitial translation of the ra w file. The Auto settings attempt to i mprove the image by ensuring that there are bright highlights, dark shadows, a good range of midtones and a dash of contrast, all depending on the actual content of the image.
___________� Si r-__________-= C__ '�� Ba � � A'-' (c:: S'h ::. ot J..CJ :) White Balance: ,, s" =----
��Ii1��l iJl�
__ __ __ __
Tone CuNe
i=: I:bu"
"
Figure 3-58. Cla rity set to 100.
Working with Adobe Camera Raw and Photoshop Lightroom I 75
Vibrance. Vibrance is another adaptive control that boosts
the color saturation in a reas that have lower saturation with lesser impact on colors that are al ready highly saturated. It will also prevent skin tones from becomi ng oversaturated, which makes it very useful for portrait images (Figure 3-59). Saturation. The Saturation slider will increase or decrease the
saturation of all colors in the image. In this respect it is similar to the Saturation slider i n the Hue/Saturation command in Photoshop though the changes it creates are more subtle. If you want a quick preview of how an image looks i n black and white, move the Saturation slider all the way to the left to - 1 00. There are ways to create a black-and-white effect that give you more flexibility and control in how to interpret the tones in the image, but using the Saturation slider (Figure 360) is a good way to get an idea i f this is a direction you want to pursue with an image. A setting of + 1 00 will i ncrease the color saturation to double the original amount.
Camera Raw and Photoshop Lightroom share a common raw conversion engine under the hood and though there are some minor functionality differences between the two, the controls are essentially the same in each program.
The Tone Curve tab
Although the Contrast slider in the Basic tab can provide a quick contrast fix, i n nearly all cases we end up ignoring that setting and using the Clarity slider for a subtle contrast boost that enhances detail, or, when we really want precise control, mov i ng over to the Tone Curve tab. Why? Because it provides us with a degree of precision and control that the standard contrast slider cannot offer. Like an RGB curve in Photoshop, the tone curve in Camera Raw places the black poin t in the lower left and the white point in the upper right. The curve maps the tonal values in an image as XY coordinates. Shadows are controlled by the lower third of the curve, midtones the middle third, and highlights the upper third. The horizontal axis represents the original tonal values of the image (input values) and the vertical axis represents the a d justed values (output). Moving the curve up will lighten and moving it down will darken. A 45-degreee diagonal line i n dicates that no changes have been made to the original values. To accomplish these adj ustments there are two tone curve i nterfaces that both allow for customized changes to the curve, but they each work a bit d i fferently.
Parametric Curve. The Parametric curve lets you adj ust specific tonal properties i n the image using the sliders for Highlights, Lights, Darks, or Shadows. The split sliders at the bottom of the curve grid let you customize the size of the area that each of the properties s l iders affects. Moving from left to right the fou r quadrants along the bottom of the graph represent the Shadows, Darks, Lights, and Highl ights (see Figure 3-6 1 ). I n the default configuration, each one represents 2 5% of the total tonal range. Move the split sliders under the graph to change the total percentage that the properties sliders a ffect. The Lights and Darks sliders will impact more of the m iddle part o f the curve while the Shadows and Highlights will be more focused on the lower left and upper right sections. In Figure 3-62 the contrast is being i ncreased slightly.
:)
-,
Highlights Lights Darks
Shadows
0 0 0 0
Input:
�
Output:
Figure 3-63. Hold down the Comm an d/Control key an d m ove your m ouse over the preview im age to see an in dicator all the curve that iden tifies the location of the image tone you are measurin g.
0---' � O-�I
If you Command/Control-click while doing this, you can place a control poi n t i n that section of the curve. To remove a poi n t from the curve, simply click it and d rag it out of the graph box.
Figure 3-61. The Parametric Curve panel in Camera Raw.
Dragging the curve up will lighte n the i mage overall while d ragging it down will d a rken it. Raising the upper section (highlights) a n d lowering the bottom section (shadows) will create a boost i n contrast. You can see this if you try out the Strong Contrast curve i n the Ton e Curve menu, or create your own S-curve (Figure 3-64) .
m
Figure 3-62. Adjustin g con trast with the parametric curve. Point Curve. The Point Cu rve functions m uch the same as a n RGB curve i n Photoshop, with the exception that it only affects the lumina nce values; you cannot access the individual color channels. A pop-up menu lets you choose between Med ium Contrast (the default), Strong Contrast, and Linear, the latter being no adj ustment to the curve at a l l (an unaltered d iagonal line). If y o u m a k e changes t o any of the preset curves the menu will change to Custom.
You can click on the curve to place a point i n a specific area. Each control point on the curve has I nput (before) and Output (after) values, which are displayed below the graph. To see where on the curve a specific image tone can be found, move your cursor over the image and hold down the Command/Control key to show a circular indicator on the curve ( Figure 3-63) .
lZl Figure 3-64. An S-curve to increase contrast. Brighter values are lightened while darker tones are darkened. ' adjustment layer-typeflexibility for rawfiles after tl1 e), have been opened in Photoshop. Double-click the Sm art Object thum bnail to open the Cam era Raw dialog and change the setting>Ior thatfile. " SI)
Working with Adobe Camera Raw and Photoshop Lightroom I 85
Ph otosh o p L i ghtro o m Getting the most from Camera R a w relies o n working between Bridge and ACR. Adobe Photoshop Lightroom represents a new approach to the raw workflow in that it combines a database-d riven image browser with the capability for enhanced raw adj ustments as wel l as output options that neither B ridge nor Camera Raw currently offer. Si nce Camera Raw is the engine under the hood of Photos hop Lightroom's Develop module, the basic theory beh i nd the adj ustments, as well as the n a mes of the controls, is the same, but there are some d i fferences. Our purpose here is not to set forth a detailed explanation of every Photoshop Lighth room feature and deta i l (there are other books that do that very wel l ) , but to give you the "big picture" of how Photoshop Lightroom fits i nto the overall landscape of digital workflow and applications that a re ta rgeted towards the digital photographer who shoots raw files.
Figure 3-90. The LibrCllY m odule in L ightroom .
The Photos hop Lightroo m interface Photoshop Lightroom is divided i n to five modules that are arranged to echo a typical workflow for worki ng with digital captures. The five modules are Libra ry, Develop, Slideshow, Print, and Web. The theory beh i n d this organization is that first you work with the collection of images (the Library) to sort, rate, and cull, then you move on to "develop" specific images, which in volves making adj ustments to the raw files, and fi nally you progress to d i fferent ways to output and present your photographs. The Library module. This a rea offers many of the same capabilities as the Bridge browser, but they are implemented much better and a re more focused on the needs of photographers; Bridge is designed to function as a hub for all of the Creative Suite programs, not all of which use image files. I n the Library pane (Figure 3-90), you can work with groups of images as Collections or organize them i n to folders and perform basic sorting, ranki ng, and culling. Keywords can be added and the image library can be filtered and searched using a range of criteria. There is even a Quick Develop palette for images that don't require major adjustments (Figure 3-9 1 ) .
Figure 3-91. In the Library m odule, lalger previews of sin gle ima ges can be viewed and a Quicl( Develop pa lette enables you to make basic a djustmen ts.
The Develop module. The Develop module offers all the functionality of Camera Raw and then some. The controls are arranged in a single scrolling panel with d i fferent sections that can be collapsed when you're not using them (Figure 3-92) . In addition to the same adj ustments found in Camera Raw there a re also features that a re specific to Lightroom, such as the ability to d rag in the histogram to adjust the tonal range (Figure 3-93) , an enhanced white balance selector tool (Figure 3-94) and the ability to compare before and a fter views of the same i mage (Figure 3-95) . The Curve i n terface in Lightroom includes add itional feedback as seen in Figure 3-96. The lighter areas on the curve reflect how fa r you can push or pull the curve before clipping the image. Additionally, a fter setting 'the cu rve you can adjust the shadow, midtone, and h ighl ight slider d i rectly under the curve i n terface to have Lightroom i nclude or exclude tonal values in the curve adjustments. When the target mode button in the upper-left corner of the cu rve a rea
86 I Chapter 3 : Scan, Develop, and Organize
is activated, you can d rag on a particular tonal region in the image preview to l ighten or darken those tonal areas. The target mode feature is also present in the HSL and Grayscale Mix controls. Just as with Camera Raw, any changes you make to files in Photoshop Light room are stored in a database or as XMP sidecar files, and the original camera file is never altered.
Figure 3-95. B efore and After comparative viewj" of the origin al image (left) and a dayjor-n ight interpretation (right) .
Figure 3-92. The Develop m odule offerj" a j'crollin g palette that containj" a varied collection of con tro ls for m odifying the im age.
Figure 3-93. In L ightroom's Develop m odule you can drag directly in the histogram to lighten or darl(en the image.
Figure 3-9 6. The Curve in Lightroom offers addition al feedbacl( and function ality. When the TClIget m ode button is activated (circled in red) , you can click and dra g on a ton e in the image to lighten or darl(en those ton al areas.
The output modules in Photoshop Lightroom-Slideshow (Figure 3-97), Print (Figure 3-98), and Web (Figures 3-99 and 3- 1 00)-all offer an impressive array ofsettings, too numerous to mention here, for presenting your images, either as onscreen slideshows, as printed output, or as Web galleries.
The Slideshow, Print, and Web modules.
For an in-depth exploration of all the featu res available in Photoshop Lightroom, see Mikkel
Figure 3-94. Adjuj"tin g white balance with the White Balance Selector tool.
Aaland's
Photoshop
Lightroom
( O ' Reilly) ,
or
Evening's
Martin
Adven ture The
Adobe
Photoshop Lightroom Book (Adobe Press).
Working with Adobe Camera Raw and Photoshop Lightroom I 87
Figure 3-97. The Slideshow m odule.
FiguI·e 3-98. The Print m odule.
Figure 3-99. The Web m odule set to the defa ult HTML ga llery.
Figure 3-100. The functiona lity of a Flash web templa te ca n be previewedfrom within L ightroom.
88 I Chapter 3: Scan, Develop, and Organize
Bridge and Cam e ra Raw, or Photosho p lightroom? Which program you decide to use (and o f course, there are others on the market) will probably be determined by how many image files you are generating, what type of capabilities and functionality you require, and whether you want to purchase (and learn) a whole new application. I f you don't shoot a lot of files and, more im portantly, if you don't shoot a lot of raw files, then Photoshop Lightroom may not be worth your time and you may be better off sticking with the still very capable combination of Bridge and Camera Raw. But i f you are a high-volume photographer who shoots a lot of raw files, then Photoshop Lightroom is most likely better suited as a complete workflow solution for you. Of course, in all of this d iscussion about working i n Camera Raw and Photoshop Lightroom, you might be wondering where Photoshop fits i nto this brave new workflow. After all, if you can do so m uch to raw files in Camera Raw and Lightroom, what is left to do in Photoshop? Working with Bridge and Camera Raw, or with Photoshop Lightroom (or similar applications) is all about making a lot of files look very good. Since the nature of these adj ustments is global, the focus should be on the overall adj ustment. Since settings can be easily applied to several images at once in both Bridge and Camera Raw or in Photoshop Lightroom, the em phasis is more on workflow.
The key part of work low is the "flow"-the movement of groups of images th rough the process to im prove them, make them look good, and, in the case of Photoshop Lightroom, to prepare them for a variety of d i fferent presentations and outpu t. In Photoshop, the focus is less on volume production and more on taking a few images and making them perfect. With capabilities such as adj u stment layers, selections, layer masks, precision corrections to specific image areas, the additions of texture and blending modes, m ultiple image composites, and a wide array of filter effects, Photoshop is the program for more individualized attention. Adobe Bridge, Camera Raw, and Photoshop Lightroom represent the production-line model of moving a lot of images from poi nt A to point B. True, the capabilities of these programs make it a very exciting and creatively stim ulating production line, but it is a production line nonetheless. Photoshop is like the artist's studio where time is spent and care is lavished on the image in creative exploration and d iscovery. Both models are essential to the d igital photographer and both can coexist, but it's important to understand where the strengths of each program lie so you can more e fficiently focus your energies on the task at hand . I n the rest o f this book we turn our attention to the artist's studio-the d igital dark room-where c reative exploration leads to serendipitous discovery and creative image making.
Bridge and Camera Raw, or Photoshop Lightroom? I 89
F i l e Pre p a rati g n Before painters can create a masterpiece with oil paints, they prepare the canvas by stretching it onto a frame a n d priming the canvas w i th multiple coats of gesso. Before master chefs can create a gourmet meal, they take the time to prepare the i ngredients: simmer the stock, julienne the vegetables, and trim the meats. Similar to the preparation traditional a rtisans do before getting to work, for photographers digital p reparation is the foundation that insu res the best quality image. While some of the preparation work in both d igital and traditional realms may be tedious, thankfully a lot of the heavy l i fting of applying the same improvements to m a n y digital files can either be automated via Photoshop actions or made easier w i th thi rd-party software. This chapter emphasizes addressing:
the i mportance of file
•
Reducing a n d Avoiding Noise
•
Developing a Sharpeni ng Strategy
•
Correcting Optical and Dimensional Distortion
•
Cropping and Cleaning Up Files
preparation by
Just as the idea of stretc h i ng canvas or peeling onions m ay not seem like the most exciting tasks, preparing a d igital file before delving i nto the finer aspects of the d igital darkroom may not sound pleasant but w i l l yield the best results i n the fin a l product. Skipping this step will resul t in lesser quality i mages and prints . So lets stir the gesso, sharpe n the k nives, a n d l e a r n t o prepare digital files t o m a k e t h e m look t h e i r v e r y best.
Chapter 4
Essential I m age Q ual ity Knowing what to look for when j udging the quality of a file is an essential skill. Identifying a problem and its cause is the first step in ferreting out the weakest l i n k in your image making process, enabling you to avoid the problem in the future while also determining the best approach to remove the problem in software. Some problems may show up only in certai n areas such as corners or individ ual channels, and this in fluences how you repai r or al leviate the problem. The two general categories we look for are photographic and digital. With photographic problems often caused by i ncorrect exposure, shutter speed, and/or poor lens construction; while i n ferior sensors and overly aggressive image processing create digital problems that are most often revealed in the del icate tonal transitions and individual image channels.
Photogra p h i c co n cerns Exposure is t h e most important technical consideration you make when holding a camera. The right exposure creates a file that requires less tonal adj ustment and although you
can recover over- or under-exposed i m ages, it is always best to start with the proper exposure as add ressed in Chapter 3. Lightening an under-exposed or dark image reveals undesired noise, and there is very little you can do to bri ng true detail back i nto those areas. If a n image is so over-exposed that the highlights a re pure white, readi ng 255, 255, 255, then there is no true detail in those a reas that can be recovered. During image capture, use the camera's histogram to j udge your exposure. I n our experience, it is best to expose slightly to the right, since darkening a n image that is approximately one-third to one-half stop l ighter than a standard exposure conceals noise while maintaining highlight deta i l . I n Adobe Photoshop, Photoshop Lightroom a n d Camera Raw, use the H istogram and I n fo palettes to j udge exposure. The best exposures are not jam med up to the left or the right side of the histogram. Underexposing Images increases noise because the u nderexposed image received less light and less signal for the camera sensor to create the i m age with, as the second cropped image in Figure 4-1 i l lustrates. Sean photographed the museum still l i fe at 3200 ISO and then accidentally underexposed the same scene by one stop.
Figure 4-1. Photographed in dim museum lighting at 3200 ISO, the properly exposed ima ge conta inssome noise, while the underexposed ima ge, which was lightened in Adobe Cam era Raw, has so m uch color nOISe tha t the I1nage IS not usa ble. Convert for Smart Filters i n Photoshop CS3, and select Filter-,> Blur-,> Surface Blur. To see the e ffect the sliders a re having, increase the Radius slider to 1 0, and then click and press inside the Threshold d ialogue box.
2.
Increase the Threshold slider to make the noise disappear. In this dialog box, it is safe to go to high because over-shooting the goal is helpful when deciding how much noise to remove.
3.
To refine the smoothing effect, click in the Radius box, and use the down a rrow to lower the radius to find the right balance. In our opinion, a Threshold setting of 8 with a Radius setting of 3 reduced the noise quite well while maintaining the details i n the lamp, as seen i n Figure 4-7.
4.
To refine the noise reduction use a soft-edged black brush to paint on the Smart Filter mask and paint over important details, such as the pink decoration on the tablecloth, the prongs of the fork, and metal parts of the lamp, as seen i n Figure 4-8. Painting with black will conceal the Surface Blur from i m pacting the details.
5.
To camouflage the color noise in large, flat surfaces such as the wall, add a new layer and change the blend mocle to Color. Sample the wall color and use a large, soft-edged brush to paint over color mottling on the wall. Option/Alt-click to change the B rush tool to the Eyedropper tool to sample a variety of desired colors to paint with.
6.
To even out the brushstrokes on the Color layer, select Filter-,> Blur-'>Gaussian B l ur with a setting o f 5 to create the resul ts seen in Figure 4-9.
FigU1'e 4-7. A higher radius applies m ore noise reduction while a higher threshold includes m ore tonal values in the noise reduction.
Figure 4-8. Painting with black on the Sm art Filter m ask allows the original image inform ation to be visible.
Figure 4-9. Paiming on a color layer effectively conceals the color ullevenness in theflat sU1faces.
Noise Reduction I 97
Working with Reduce Noise The Reduce Noise filter allows for greater control over noise reduction, though it is a very slow filter to work with and the controls are not named i ntuitively. I n its defense the Reduce Noise filter reduces both luminance and the tougher color noise well (as illustrated in Figure 4- 1 0), and it includes a "save settings" option to save and reload your commonly used settings.
JEJ ch4_fash ion.jpg
Figure 4-11. To understan d how the Reduce Noisefilter works it ,�. useful to zero out a ll sliden an d a dju st them one at a time.
Strength: The strength slider controls how much luminance noise is removed. I n terestingly, even with a zero setting you can a d j ust the Reduce Color Noise slider to remove color mottli ng.
QRomoveJf'(C Af1'/;U:\
Figure 4-10. Photograp hed a t 1000 ISO, there i s color n oise in the sha dows and un desired texture on the m odel�' back. A fter Noise reduction thefile looks much bette}: f) .\/'IIA' Ik( I(clmal/
To speed up the Reduce Noise filter and make
Figure 4-12. The Strength slida:
Preserve Details: Use this slider to maintain details, which on a noisy file can be a n exercise i n frustration as fine detail and noise a re handled in a similar fashion. For best results, work back a n d forth between the Strength slider and the Preserve Details slider.
changing the settings more responsive, make s u re that the Preview checkbox is unchecked. This tells
E:K3 �
Photoshop to apply the filter only to the p review in
OPr.�I.w
the Reduce N oise interface, not to the actual file. For a q uick before and after view, click and p ress the mouse button
on
the Reduce Noise preview to
St,.ngth:
mouse to see the filter ap plied.
Pruorv. o.Ulio: rso--l X •
R•.,uu Coler
•
I n Basic mode, the filter i n terface offers five controls, as seen in Figure 4- 1 1 through Figure 4- 1 7. Figure 4-13. The Preserve Deta ils slidel:
98 I Chapter 4: Fi Ie Preparation
rr--l
•
see before noise reduction, and then release the
NOI'.:ro-i"
Reduce Color Noise: The most effective slider of all, this reduces the color noise. We prefer not to i ncrease the slider above 50% as colors have a tendency to become too desaturated. If you require a higher setting, a fter using the Reduce Noise fil ter add a H ue/Saturation adj ustment layer and increase the saturation by 1 0%.
Figure 4-1 6. Save Settings feature.
Figure 4-14. The Reduce Color Noise slider.
Remove J PEG artifacts: This settin g is ideal for reducing the small, square artifacts that lowend digital cameras, cell phones, or overzealous compression can add to a file. Check Remove JPEG Artifacts and adjust the Strength and Reduce Color Noise sliders (Figure 4- 1 8) to gen tly soften the artifacts away.
Sharpen Details: Use this slider with caution, as it can add horrific artifacts and c reate more gritty problems than it solves. I t is best to leave Sharpen Details at zero and to apply sharpening separately in the following Capture Sharpening phase of file preparation. Working i n Advanced Mode: This mode enables you to apply additional lumina nce noise reduction and control the details on the i ndividual chan nels. Keep in mind that these settings are applied on top of the settings applied to the enti re file. In most cases, the blue channel needs more noise reduction. Figure 4-17. Remove jPEG artifacts. @lu). Q AdvInc:."
: �a
Figure 4-15. Working in A dvanced Mode.
Save Settings: With so many sliders, settings, and variations, we are very glad that there i s a save settings feature. Click the tiny floppy d isk icon to save worthwhile settings with useful names.
8
Figure 4-18.
10M
j.,/;,;
�
.\1
IlliI
� Preview
X,
Figure 4-27. The Find Edgesfilter out/ines the image contours, and the Levels enables y ou to increase the con trast needed to make an effective mas/(.
Noise Reduction 1 1 03
Using the edge mask
Tonal masking
1.
Duplicate the Background layer. Select-Load Selection, and use the Channel pull-down menu to select "Edge Mask."
2.
At the bottom of the Layers palette, click the layer mask button to transfer the active selection to the layer mask. Before applying noise red uction, be sure to click the d uplicated layer icon to make sure it is active (rather than the layer mask.)
3.
Use any of the noise reduction methods described up to this point; in this example we used the Reduce Noise filter as seen in Figure 4-28. Upon close inspection we were very pleased to see how well the fi ne details of the pine trees were maintained while the clouds were cleaned up.
Noise most often lurk s i n the shadows. A tonal mask e n ables you to reduce noise in the shadows while not a ffecting the midtones or highlight areas as seen in the comparison i n Figure 4-29.
To view a layer without the layer mask, Shift-click the layer mask icon. Shift-click again to turn the mask back on.
� Ol'f,vitw
Settings:
!Oefilult
':I � a
SUtnglh:{2l
FigU1·e 4-29. Using a tonal m ask to pro tect highlights allows for more effective no ise reductio n in the shadows. (,, " I-"
� ch4_leaves.jpg
FigU1·e 4-28. Applying the noise reduction through the edge m ask maintainsfine detail.
1 04 I Chapter 4: Fi Ie Preparation
1.
Click the Channels palette and view each channel to find the one with the greatest contrast. Duplicate the channel by d ragging it down to the New Channel button.
2.
Select Image-Adj ustments-Levels, and increase the shadow point to at least 50 and the highlight slider to approximately 200, as seen in Figure 4-30. This creates a very contrasty i m age.
3.
Use the Gaussian Blur fi lter on the channel mask with a setting of 5 or higher to soften the transition from the shadow to midtone areas.
4.
Return to the Layers palette, select the background layer, and then select Filter-Convert for Smart Filters.
5.
Use the Noise Reduction filter to remove the noise, as seen in Figure 4-3 l .
6.
Click the Smart Filter mask, and then choose Select-> Load Selection. Use the Source Channel pull-down menu to select Alpha 4, or use the key command Cmd + Option/Ctrl + Al t + 4 to activate the alpha channel as a selection.
7.
Choose Edit->Fill Contents Black to block the noise reduction from a ffecting the less noisy midtones and highlights and creating the fin a l results seen i n Figure 4-32.
� � � �
f20ifl ( Opt;O",... ) -------
1/I/l�1
� Preview
X-
Xl
Figure 4-32. The light areas o/the Smart Filter mask are where the greatest noise reduction will take place in the image.
Xl
Use tonal masks to control where tonal, color, and sharpening enhancements take place. By inverting
X3
the mask you can adjust either the lighter or darker areas of the image without affecting the other.
X4
Avo i d i n g N o i se Before Pressi n g t h e S h utte r
Figure 4-30. C" eating a high-contrast tonal mask.
hi " 8 Strtngth:rJ Pren� Deulls:
e
lI.eduu_ Color Noise:
S
As you can see, a great many techniques and products are available for reducing noise. After w restl ing with a wide va riety of noisy files, we d o our very best to avoid noise whenever possible, using these guidelines: 1.
Avoid usi ng the higher ISO settings u nless absol utely necessary due to low l ight, to increase depth of field or to freeze motion.
2.
With digital cameras, expose to the right by approximately half an f-stop. Darkening a light file conceals noise, while l ightening a d a rker file reveals nOise.
3.
Use faster lenses a nd/or embrace a shallow depth of field a n d motion blur aesthetic.
e
roI" fin"
Figure 4-31. The shadow areas usua lly conta in the m ost noise.
Noise Reduction 1 1 05
4.
Take a serious look at working with i mage stabilization lenses, which enable you to shoot at slower shutter speeds or lower ISO settings.
I n p u t Sharpe n i n g
5.
Consider buying a new digital camera as the newer ones have better analog-to-digital conversions and much better l ight sensitiv ity performance-both of which create higher quality and cleaner files.
6.
Keep your camera cool as heat creates non-signal information. Leaving your digital camera in a hot car is a surefire way to create noisy files. Thankfully, once the camera cools down it will perform as before. We treat our digital cameras like film, protecting them from extreme temperature shifts, humidity, dust, and sand.
The key to applying delicate i n put sharpening lies i n making an inverted edge mask that enables you to sha rpen only the image edges as seen in Figure 4-33.
Shar p e n i ng Process After applying noise reduction a n d flattening the i mage, it's time to en hance the detail of the image with the fi rst step of a three-step sharpening process. •
•
•
I n put or capture sharpening: Offsets the softness that lenses, a nti-aliasing filters in digital cameras, and scanner optics add to an image. This fi rst pass of sharpening needs to be done th rough an edge mask and should ideally sharpen only the midtones of an image and leave the shadows and highlights untouched . Creative or interpretive sha rpen ing: Incl udes sha rpen ing and blurring techniques to enha nce image focus, depth of field, and atmospheric effects. Creative sharpening is applied a fter global and local tone and color correction and enables you to enhance the subjective and emotional impact of an image. This topic is addressed in greater detail i n Chapter 1 0. Output or print sharpening: Each type of output, including offset and i n kj et printing and onscreen output for the Web or for presentations, adds softness. Output sha rpening takes the dot gain or dither patterns i nto account and creates a c risper image for reproduction. Sharpening for i n kjet printing is addressed in the online section "The Print."
The folks over at PixelGenius have put a lot of thought,
testing ,
and evaluation into creating
PhotoKit Sharpener, a Photoshop plug-in that uses a three-phase sharpening method to help users improve their images quickly and easily. PhotoKit Sharpener
considers
the
image
source ,
file
character, and final output to produce excellent results.
Figure 4-33. Cal'eful input sha/"jJening tightens up the image edges to create a delicately crisp image.
1 06 I Chapter 4: Fi Ie Prep aration
Ii) KF
-1EJ
ch4_caution.jpg
-1EJ
ch4_inputsharpen.atn
1.
In the Channels palette, duplicate the green channel by dragging it to the New Channel button.
2.
Select Image-> Adjustment->Curves and, as shown in Figure 4-34, reduce the highlights to add a wash of gray and to deepen the shadows, which will later protect the shadows from being sharpened.
Amount: � "
ItadiUS:[.7S
e
[ pillels
e
3.
Select Filter->Stylize->Find Edges, followed by Filter->Blur->Gaussian Blur of .5 and Image-> Ad justments->Invert, to create the mask seen in Figure 4-35.
4.
Return to the composite channel. In the Layers palette, use Cmd-Option/Ctrl- Alt+4 to activate the edge mask, and then choose Layer-> New->Layer via Copy (or simply press Cmd/Ctrl+ J).
5.
Choose Filter->Sharpen->Unsharp Mask, and use relatively low values of 75/.75/0 to add crispness to the edges. Change the layer blend mode to Luminosity to avoid aggravating any possible color artifacts (Figure
4-36).
Figure 4-35. After shalpening the edges, refine the effect with Luminosity blend mode and layer opacity.
After apply i n g captu re sharpe n i n g, flatten the layers, delete the alpha mask, and use File->Save As to save a work i n g copy of t h e s h arpened file. For add itional i nformation on i mage s harp e n i n g, please see Real World Image Sharpening by B ruce Fraser.
Input Sharpening with Adobe Camera Raw Ch;l,nnet: j Green copy
�Prll!vll!w
XI
Xl
�I 0..lJZI C Show Cllppmg 1255 8 CUM OispiayOpllons -----
X'
One of the best aspects of working with raw files is that as the software is updated and improved, you can return to your image collection and reprocess images with more control and higher quality. With the release of Adobe Camera Raw 4.1 for Photoshop C S3, Adobe introduced very impressive raw sharpening with edge-masking controls that sharpens only the luminance information. In our opinion, the sharpening controls in Adobe Camera Raw 4.1 (and higher) and Photoshop Lightroom (1.1 and higher) usurp the need to apply global Input Sharpening in Photoshop. The image seen in Figure 4-36 shows how subtle and effective the ACR sharpening controls are.
FigU1'e 4-34. Flattening the tonality will protect highlights and shadowsfrom being shmpened.
S harpen i ng Process 1 1 07
"' ---
G: --B: ---
I
f/9 ISO 100
1/1605 24-105@10Smm
Sharpening
t���__________�o
__
m ,"" '
Radius
0.5
C Masking
C
No sharpeni n g
I B� ili iilr�J . £!l R' ---
f/9
�; :=:
ISO 100
1/160s
24-I05@105 mm
Dewl
Shllrptnlng 25 �
�___________
IO __
'2 5 c M.:uking
C
Default s harpeni n g f/9 150100
111605 24-10S!JI05mm
�lA'lrIli11iil.if li'l jf Dtlall
Sharpening Amount Radius
15 c
DetaU c Masking
70 c
I mage-appropriate s h arpe n i n g
Figure 4-36. Adobe Camera Raw provides controlsfor the strength and dimension ofthe applied sharpening.
Detail When adjusti n g s harpe n i n g i n Adobe Camera Raw, make s u re yo u are view i n g the i mage at 100% to see the effect.
The controls in Adobe Camera Raw are:
Masking
Amount
From 0 to 150, with higher settings applying a stronger e ffect. The de fault for non-raw images is 0 and for raw images is 25. We find the Amount to be the least important slider o f the four, as the remaining sliders o ffer finer controls. Start with a high Amount, such as 100, ad just the remaining sliders, and then return to the Amount slider to refine the overall e ffect of the sharpening. Radius
Controls how many pixels (from 0.5 to 3 pixels) on either side o f the edge are included in the sharpening.
1 08 I Chapter 4:
The Detail slider controls how fine image areas are treated. Low settings apply a halo dampening on the sharpening edge set by the Radius slider and create a very tight fine-etched look. Higher settings make the sharpening perform more like the traditional Unsharp Mask filter.
Fi Ie
Preparation
Creates an edge mask, just as the edge mask built in Photoshop controls which image edges are included in the sharpening. Since this edge mask is created on the fly and can slow down the responsiveness o f Adobe Camera Raw, we suggest leaving the mask setting at o (zero) while ad justing the previous three sliders and then adjust the mask as a final refinement. To work with the sharpening e ffectively, make sure to view the image at 100% and also take advantage o f the hidden preview. Press Option/ Alt to view the ef fect that each slider is having as seen in Figure 4-37.
The focal length of the lenses we use impacts the flattening process. Longer lenses create a forced perspective in which the background is made larger in relationship to the foreground, while wide-angle lenses cause the background to recede and seem smaller. Wider lenses also have a tendency to introduce bowing of vertical and horizontal lines that are close to the edge o f the frame. In Photoshop, you cannot change focal length, photographic perspective, or point o f view, but you can correct some lens shortcomings. d
c
The optical issues to correct include: Image Levelness
The slightest misalignment caused by not positioning the camera per fectly perpendicular and level to the subject can cause the image to be slightly o ff-axis to the horizontal or vertical axis. Level the image first to make it easier to see other possible optical problems. Keystoning
a. S harpened Image b. Amount L u m i n ance i nformation c. Radi u s effect edge d . Detail preview e. Mask
Figm'e 4-37. OptionlALt-cliclCing each sLider controL reveals the intemaL effect being appLied.
Unless syou're working with a view camera or a shi ft/tilt lens, which allows control o f the alignment o f subject, lens plane, and image plane, the vertical or horizontal lines o f the image will converge and become narrower the further away they are from the camera. O f course, the receding lines can be used for creative e f fect, but when photographing subjects that should be straight, such as architecture or product shots, proper alignment o f horizontal and vertical lines are the mark o f a pro fessional photographer. Barrel and Pincushion Distortion
To study and adj u st the affect of the Camera Raw sharpe n i n g, open the raw file as a S mart O bject.
The sharpening in Adobe Camera Raw does not replace additional creative and output sharpening and is best used to enhance an image, not to interpret it. Take the time to adjust each slider so that the image looks good at 100% view, not crunchy as older sharpening procedures called for. Once you've found settings that work well for your camera, lens, and image types it makes sense to save the settings in the Presets section o f Adobe Camera Raw.
Co rrecti ng Opti cal Disto rt i o n and Pe rspective We live in a three-dimensional world made up o f width, height, and depth, also called the X, Y, and Z axis. Photography is inherently a two-dimensional process in which the three dimensions are flattened onto a 2D surface when we press the shutter. It is at this point in the process of flattening three dimensions down to two that undesired distortions and perspective changes are created.
Barrel distortion is a bowing out and pincushion distortion is a pinching in o f image edges. Both o f these distortions are more exaggerated with wider, lower quality lenses and zoom lenses on inexpensive point and-shoot cameras. Vignetting and Flare
Vignetting is the darkening o f the image corners, which may or may not bother you. Vignetting is best fixed in the raw acquire, while removing vignetting from a film scan or flare is a retouching process and can be concealed in the Spotting and Clean-up phase. Removing flare is a challenge as it not only leaves telltale sun splotches, but it also reduces image contrast. Chromatic Aberration
Color fringing usually found near the image frame and more visible on high-contrast image areas. Lower quality and wider angle lenses have a greater propensity to create chromatic aberration. It is best to correct for chromatic aberration in the raw acquire phase o f the digital workflow and before applying input sharpen ing.
Correcting Optical Distortion and Perspective 11 09
The tools and techniques used to correct optical shortcom ings include: Raw acquire software to remove chromat ic aberration, vignetting, and straightening images as addressed in Chapter 3; Photoshop Lens Correction filter to correct distortion and keystoning; transform controls and the Crop tool to correct perspective and d istortion; ad justment layers and retouching tools to remove vignetting and flare; and third-party applications such as DxO to process and correct a wide variety of lens shortcomings in the raw acquire, all of which are addressed here.
The lens Correction Filter The Lens Correction filter is an ex cellent tool to straighten, rotate, remove d istortion, and correct vertical or hor izontal perspective. The ab ility to apply a variety of corrections while seeing how each one improves the image is the strength of the Lens Correction fil ter. Working efficiently with the Lens Correction filter 1.
Before selecting Fi lter->D istort->Lens Correction, imagine the subject of the photograph imposed on a cube: a single face of the cube when the photograph was shot in one-point perspective, or an angled cube when photographed in two-point perspective, as seen in Figure 4-38. For us, imagining the subject on a cube makes it easier to see which lines need to be straightened or where d istortion needs to be removed.
2.
Set the grid to a useful size and color. The defaults of 16 and gray are usually too small and too subtle to be very helpful. We use 32 w ith a red gr id . Use the Move Grid tool to position the grid along important l ines such as the edge of buildings or along the horizon line to give you a reference for what should be straight and whether the corrections are improving the image .
3.
Start the correction session by straightening the subject with the Straighten tool by dragging along a vert ical or horizontal l ine. If the subject isn't straight, it will be more difficult to use the other correction features.
Figure 4-38. The same abandoned security house photographed in one- and two-point perspective. (0 f... F On the right side of the Lens Correction interface, work from top to bottom. Keep in mind that correcting one flaw may make others more apparent, wh ich calls for a flexible working back and forth between the straighten, remove distortion, and transform: Remove Distortion
Corrects bowing or pinching (officially called barreL distortion) when the image bows outward and pincushioning when the image edges pinch in. Chromatic Aberration
Removes color fringing most often found near the edges of images in high-contrast areas. If you work with a digital camera that can capture in the raw format, it is better to remove chromatic aberration with Adobe Camera Raw or Photoshop Lightroom.
1 1 0 I Chapter 4: Fi Ie Preparation
Vignette
Lightens dark corners that were created by large lens shades or zoom lenses. Vignette can also darken corners for a creative effect.
1.
Duplicate the background layer.
2.
Select Filter�Distort�Lens Correction. Check the Show Grid box for assistance in determining the lines that need to be straight. Place the mouse cursor in the Remove Distortion box, and hold Shift while tapping the up arrow key to remove distortion in one-step increments. In this example, we used a +4.00 to reduce the bowing.
3.
Next, use the Vignetting slider to lighten the corners.
4.
The Transform sliders can dramatically correct keystoning. Sliding the Vertical Perspective to + 12 straightens the wall. Tweaking the Horizontal Perspective to -\ brings the left side of the building to the foreground ever so slightly, as seen in Figure 4-40.
5.
After accepting the Lens Correction, use the Crop tool to crop out the extraneous edges.
Set Lens Default
Useful when using prime lenses (not zoom lenses). After correcting an image, click Set Lens Default to save the corrections for the next time an image is opened that was photographed with that same lens. Transform
Corrects perspective to straighten up buildings. Edge
As the image is scaled and corrected, transparent areas are created. The options to conceal this are Transparency, Edge Extension (not recommended), and Background Color. We recommend either scaling to increase the image to hide the transparent edges or ignoring this feature and using the Crop tool afterwards.
�
i[-:� .
[ =:::iT]
Correcting lens distortion
When we use different focal length lenses, we often trade off an optical distortion for the ability to get the whole image within our viewfinder. This is very noticeable when a wide-angle lens is used. As the lens takes in a wider view, a noticeable bend is created in straight lines. The image of a contemporary building, taken with a 20mm lens, is distorted, which is especially visible in the vertical lines near the edge of the frame, as seen in Figure 4-39. With just a few adjustments inside the Lens Correction filter, the lines are straightened and the bowing is removed.
�
.
-':'� .I � ,� __t
-�'.- . .
--
-0
...-
....
.
= �.
ch4_gl assbuilding.jpg Figure 4-40. Fine-tuning and adjusting the building was achieved by removing distortion to straighten the building right out. Correcting architectural perspective
If you use a 35 mm film or digital SLR camera to photograph a building, then you'll notice that the building looks as if it's falling backwards. But when you look at the actual scene, the building isn't tilting at all. Our visual system corrects the perspective just as a view camera does, by aligning vertical and horizontal lines to make the building look as straight as it really is, as shown in the before and after comparison in Figure 4-41.
�
ch4_convent.jpg
Figure 4-39. Photographing with a 20 mm le11S caused the wall to bow inwards. After using the Lem Correction filtel; the wall is leJS distorted. C f( I:
Correcting Optical Distortion and Perspective I 1 1 1
Duplicate the image layer before entering the Len s Co rrection filter in order to u se the Edit> Transform control s afterwards. S mart Filter layers can not be transformed with skew or distort, which is why we d u p l icate the layer rather than convert for S mart Filter when working with the Lens Correction filter.
---- ,---. -.. . ---.
-(1)
,.�j. �..
•..
..
Figure 4-42. Do as mud) correction as you can in the Lens Correction filter intel/ace. One essential control that the Lens Correction filter does not have is the ability to pull and push individual image corners into position. This is exactly why we recommend duplicating the image layer before entering the Lens Correction filter. After using Lens Correction, turn the visibility of the corrected layer on and off. If you feel that the image still requires adjustment, use Edit ..... Free Transform to pull and push corners into place, as seen in Figure 4-43, to create the final image.
Control lright mouse b u tton on the transform anchor points to access all of the transfo rmation controls.
Figure 4-41. Before and after correcting the optical and lens distortion. iD I\./:"
Although it's always better to create the best image in front of the lens and not rely on Photoshop to fix mistakes, Katrin doesn't travel with a view camera, which means she relies on the Transform and Lens Correction features to correct images that curve, bend, fall backwards, and generally are out of alignment. The Lens Correction filter interface seen in Figure 4-42 shows how she straightened the top of the building, removed distortion, and brought the building up via the Vertical Perspective slider.
1 1 2 I Chapter 4: File Preparation
Figul'e 4-43. Use Free Tram/ann to refine individual image planes.
When p hotogra p h i n g a scene with the i ntent of correcti n g it with the Len s Correction filter, frame the scene in the viewfi n der, and then take a few steps back to provide the s pace needed to correct and scale the i m age.
Cropping and Cleaning Up Images At first glance, cropping may seem like a rather mundane and self-explanatory process. How hard could it really be? Click, hold, drag, and accept the crop. Cropping an image is actually a critical aspect of image composition. The essential question to consider is whether to crop early in the digital darkroom process or later on. The advantages of cropping early are you'll be working with a smaller file, and all image enhancements such as edge burning or adding neutral density filter effects will be accurate. Imagine if you took the time to darken down the edges of the image perfectly and then cropped the file. All of the burning work would be
discarded and would have to be redone. The advantage of cropping late into the digital darkroom process is that you maintain all of the image information. Quite honestly, we both concentrate on composing the image in the viewfinder exactly the way we want the final image to be printed and use the Crop tool only to improve the image, not as an excuse for poor image composition. The Crop tool can be used to recompose and straighten images and correct perspective. Before you grab the Crop tool and start slicing away at your images, always check: •
•
•
The Options settings: Make sure that they are zeroed out by clicking the Clear button to clear settings you may have used in a previous digital darkroom session. The Layer palette: Convert the Background layer to a standard layer by double-clicking it and renaming it or duplicating the Background layer. This enables you to hide the cropped image information versus deleting the image information. Tap F: Full screen mode provides better access to the Crop tool handles when refining composition and/or mcreasmg canvas size. .
.
.
Third-Party Optical Correction It can be rather discouraging to realize how much distortion, vignetting, and edge softening even very good lenses may add to an image. In fact, until the distortion is
mapping and analysis of the most commonly used lenses, Dx O Optics Pro corrects for lens distortion, vignetting, and overall and corner softness and enhances exposure
removed you often don't even realize how contorted your images may be. One elegant solution to correct a variety of optical defects is DxO Optics Pro. Katrin has tested this product and uses it in her own work, especially when she needs to correct images taken with zoom and wide-angle lenses with d SLR cameras. Based on a highly scientific
see the same image-once processed via Adobe Camera Raw and once with Dx O Optics Pro in which Dx O Optics Pro corrected the lens distortion, vignetting, and uneven softness quickly, easily, and effectively.
and color in the raw acquire process. In Figure 4-44 you
Correcti ng Optical Distortion and Perspective 1 1 1 3
Recomposing images
Ideally, each i m age would be perfectly composed i n the camera, but m any times the fleeting moment requ ires quick reactions or mod i fications i f, upon see i ng the i m age in the raw acquire or i n Photoshop, Katrin cropped the photograph i n Figure 4-45 i n Adobe C amera Raw to emphasize the freshness of green growth i n the lav a fields.
Crapping Tips Before cropping, double-click on the background layer to con vert it to a standard l ayer. After sett i ng the crop but before accepting i t m ake sure to cli ck 'Hide' i n the options bar to hide the extraneous i m age i n formation w ithout deleting i t . o
o
o
When croppi ng to recompose an i m age, work with zeroed-out Crop tool settings that e n able you to draw the crop freely. When working close to the i m age edge, press Cmd/Ctrl to turn off the s nap-to-edge behav ior of the Crop tool. Use the Crop tool to extend C an v as Size by d ragging the crop handles outside the existing i m age frame and accepting the crop.
Straightening and correcting perspective
As simple as it sounds, straighten i ng i m ages takes care and practice. Learn to take advantage of Photoshop's Guides, Grids, and Measure tool to m ake short order of crooked pictures, creating the results seen i n Figure 4-46.
'1l
ch4_bluewater.jpg
Figure 4-45. The tighter crop emphasizes the horizontal composition ofthe green/ems growing through the black lava. New Adjustment Layer-">Levels. You can also click the Adj ustment Layer button at the bottom of the layers palette, and then select Levels from the d rop-down menu. I n a grayscale image, or when used for overall adj ustments in a n RGB file, the Levels d ialog really controls only th ree aspects of the image: the white point, the black point, and the
1 34 I Chapter 5 : Tone and Contrast
midtone val ue. The main controls a re two sets of sliders, the I nput Levels and Output Levels, and correspond i ng numeric fields for the slider values (Figure 5- 1 0). There a re also three eyedropper tools for assigning precise val ues for h ighlights, shadows, and midtones. Depending on the adj ustments you make to the endpoints, i ncreases or decreases in contrast can also be achieved with Levels, though you cannot emphasize one area of the tonal range over another as you can with Curves. Midtone or Gamma slider Channel selector Histogram
Auto Adj ustment option Save cu rrent Levels settings
Load previously saved
Setting the black poi nt. The I n put shadow slider sets the black poin t (level 0) for the image. As you move the slider to the right, the d arkest tones will gradually become a solid black. Since the mid tone slider is moving at the same time (to always reflect the current m idtone poin t) the rest of the image will also become darker.
When you change the position of the I nput shadow slider, any tonal values on the h istogram that a re to the left of the slider will be remapped to 0, which is a total black with no detail. If there a re any subtle shadow details in those darker areas, you will be losing that deta i l . The term most often used to describe what happens when tonal detail is adj usted to either a total black or a total white is "clipping" (think of a pai r of scissors clipping off a part of the tonal range). You cannot make the shadow tones lighter with the I nput black poin t slider; you can only make them darker (Figure 5- 1 1 ). Figures 5 - 1 2 th rough 5- 1 4 show the results of a Levels adjustment. Clipped shadows shown in b l u e
( ( Output S hadow slider I n p ut Shadow slider (black point)
( (
O utput H i g h l ight slider
(
Eyedroppers for setting shadow, m idtone, h ighl ight
The I np u t Level s
A s mentioned previously, you can o n l y d irectly affect th ree tonal points in Levels, and the I n put Levels provide the primary means to do this. Most of the work you a re l i kely to do in Levels will be accom plished with these controls.
Load ... Save ... Auto
) ) ) ) )
( Options . . . )
1/1 .1' 1"'1
I n put H ig h light slider (wh ite point)
Figure 5-10. The Levels dialog in Photoshop CS3.
OK Cancel
•
E:=J
0
�
� Preview
Figure 5-1 1. /n the illustration above, the black point slider has been set to 42, meaning that all values of 42 or lower will be remapped to level 0, a total black.
I n simple terms, the effects of adj usting the t h ree I n put Levels sliders can be described this way: moving the black point slider to the right darkens the shadows by setting a new black poin t, mov i ng the wh ite point slider left lightens the h ighlights by designating a new white point. The midtone slider reassigns the m iddle tonal value i n the image. Movi ng it to the left will lighten the image, and moving it to the right will da rken the i m age. That's the quick explanation. Of course, there is more to it than that, and understanding how the tonal data is being affected is i m portant to maintaining the tonal quality of your file.
Figure 5-12. The Bodie church image before any adjustments.
� Sf)
Global I mage Improvements 1 1 35
( (
1�5 I Output Levels:
Figure 5-13. The Bodie church image, after the Levels adjustment seen in Figure 5-1 1. Detail in the darlcest shadows has been clipped to black·
6K Cancel
) )
( (
Load .. .
(
Auto
)
(
Options . . .
)
Save . . .
) )
[Zl�I/1
Before
After
Figure 5-15. By adjusting the Input Levels blade point slider to where the shadow histogmm detail starts (in this case, level 12), the deepest shadows become a richer black without clipping any important details. Setting the white
point. The I nput high l ight slider determines the white point (level 255) for the image. As this slider is moved to the left, bright highlights will start to become total white. Just as with the black point slider, the midtone point will also move along with the white point slider, and this will cause the rest of the image to become brighter.
Figure 5-1 4. As a rej'ult ofthe Levels adjustment shown in Figw'e 5-1 1. Significant shadow detail has been lost in the dar/eel' shadow tones.
I f you want to reta i n all of the i m portant shadow detail i n an i mage, you should not m o v e the black poi nt slider past where the histogram detail (the mountain shape) starts (Figure 5- 1 5).
1 36 I Chapter
5:
Tone and Contrast
When the position of the I n put highlight slider is changed , any tonal data on the h istogram that is to the right of the slider will be clipped to level 255, which is total white with no detail. A common photographic term to describe h ighlights suffering from such a malady i s "blown out." You cannot make the highlight tones da rker with the I n put w h i te point slider; you can only m ake them brighter. See Figures 5- 1 6, 5- 1 7, and 5 - 1 8 for examples.
Original midton e
C l i p ped h ighl ights shown in blue
( ( ( (
10
6
1
) )
) Save . . . )
load . . .
(
Auto
)
(
Options ...
)
1/1"'1"'1
Output level s : •
OK Cancel
[255l
� Preview
levels clipping display. When making changes to the black and-wh ite points of an image, it i s useful to see exactly what detail may be a ffected by your adjustments. To see a clipping preview, hold down the Option/Alt key while clicking either the black point or the white point slider. As you move the sliders to the left or the right you will see the following: •
•
When moving the w h i te point slider, the image will turn black, and, as the slider is moved to the left, white will begin to appear wherever tones a re cl i pped to 255 (Figure 5- 1 9). When mov i ng the black poin t slider, the image will turn white, and, as the slider i s moved to the right, black will begin to appear wherever tonal val ues a re being clipped to level 0 (Figure 5-20).
Figure 5-16. In the illustration above, the white point slider has been set to 194, meaning that all values of 1 94 or higher will be remapped to level 255, a total white. The adjustment above would result in significant highlIght clipping.
Figure 5-1 9. The Levels clipping display shows white where highlight values are clipped to 255. In the Histogram palette you can see the clipping show as the adjusted histogram data is cut off on thefar right side (the original 11istogram is shown in red).
Figure 5-1 7. The Bodie street image, before the Levels adjustment seen in Figure 5-16. If)
Figure 5-20. The Levels clipping display shows white where shadow values are clipped to O. In the Histogram palette you can see the clipping show as the adjusted histogram data is cut offon thefar left side (the original histogram is shown in red). Figure 5-18. The Bodie street image, after the Levels adjustment seen ill Figure 5-1 6. Detail ill the brightest highlights has been clipped to white.
In a color photo, the clipping display uses different colors to i ndicate clipping in the various color channels or in all the channels. We will cover using Levels on color images in Chapter 7.
Global Image I mprovements 1 1 37
The Midton e slider. The Midtone or Gamma slider determines where the exact m iddle tonal value is (50% gray, level 1 28, or Zone V). Moving it to the left will l ighten the image because that changes a current value that is darker than 1 28 to be the new level 1 28. Moving it to the right will darken the i m age because you are adj usting a current tone that is lighter than 1 2 8 to be the new level 1 28. Because i t adj usts overall image brightness without a ffecting either the black or the white point, this slider is also sometimes referred to this as the Brightness slider (Figure 5-2 1 ).
The Output Level s
'1J
ch5_Bod ie_Bed.jpg
As the name suggests, Output Levels are i ntended for making endpoint adjustments to compensate for the l i m itations of a specific type o f output. The slide rs do the exact opposite of the corresponding controls for the Input Levels. Moving the shadow slider to the right lightens the shadow tones, while mov i ng the h ighlight slider to the left will darken the highlights and make them duller.
Figure 5-21. The Input Levels midtone slider sets the midtone point and adjusts the overall brightness ofthe image without affecting the blacl( or white point. «) Sf)
E (
Ott
Cancel
) )
load . . . Save. . .
Auto
( Options ... )
B:
2 5 5 / 240 2 5 5 / 240 2 5 5 / 240
Figure 5-22. Output Levels, intended primarily to compensatefor the more limited tonal range ofan output device, wilL lighten blacks and darl(en whites in an image, but they cannot restore shadow or highlight detail that was never recorded in the original exposure.
1 38 I Chapter 5 : Tone and Contrast
When you set the shadow Output levels to 20, for example, any tone that is currently at level 0 (black) will be adj usted to level 20. All other tonal values will be adj usted proportionately, with the end result being a lightening of the image and a decrease in contrast, especially in the darkest shadow areas (Figure 5-22). If you were to move the Output highlight slider until the number above it read 240, then any current tone with a value of 255 (white) will be darkened to level 240. This will darken the highlights and reduce contrast.
Although the O utput Level s lets yo u remap a total black to a l ighter tone or a blown-out h ig h l ight to a s l ightly darker tone, it w i l l not be able to restore detail that was never captu red in the original expos u re or scan. I f i mage deta i l in such areas was never recorded, then it's gone for good.
The Output Levels compress the tonal values so that they fit within a narrower range, essentially reducing the number of tones so that the i mage can be printed within the more limited parameters of a given output medium. You would use these controls to ensure that deep s hadows don't block up to a total black or that a bright highlight with detail doesn't print as pure paper white. We usually prefer to make such adj ustments by mov i ng the e ndpoints i n Curves, because it gives us more control over how the adjustment a ffects the mid tones.
I f you do decide to use the eyedropper to set the endpoi nts in a grayscale fi le, you s h o u l d be aware that the defau lt values are 0 (total black) for the shadows and 255 (wh ite with no detai l) for the h igh l ights. As t h i s w i l l introduce c l i pp i n g i nto the shadows a n d h igh l ights, dou ble-c l i ck each eyed ropper to set a new target val u e in the color p i cke r. Safe RG B values that do not cause c l i p p i n g wou l d b e 10, 1 0, 1 0 f o r the s h adows, and 248, 248, 248 for the h i g h l ights.
Auto correction options
Just above the eyedroppers are buttons labeled Auto and Options. The Auto button applies a n automatic Levels correction based on the location of the brightest and da rkest tones in the i m age. With a grayscale file, the default setting causes the shadows and h ighl ights to be cli pped by 0. 1 0%. This action is esse n tially the same t h i ng as moving the endpoi nt sliders to w here the h istogram data begins, except i n this case a tiny amount of clipping takes place (Figure 523). Because this is a canned formula a d j ustment, and one that introduces minor clippi ng, we generally don't think much of i t and don't recommen d it. Clicking the Options button enables you to alter the default clipping values (Figure 5-24). With an RG B image, the Options button provides more choices, which can sometimes be usefu l .
The Levels eyedroppers
The eyedroppers i n the Levels dialog box can be used to target specific tones i n the image and change them to a precise tonal value. Typically, these are most often used for targeting the white and black poi n ts for a certai n output device. The m iddle G ray eyedropper, which is used to help neutralize color casts i n a reas that should be neutral, is unavailable when you are working on an image in G rayscale mode. Though they can be used to set the endpoints for grayscale images, moving the endpoint sliders to where the histogram data starts on either end accomplishes the same thi ng, so it's debatable whether or not you even need to use the eyedroppers for grayscale files. The one situation where they do come i n handy is if you need to precisely target the endpoint values for the l imitations of an output process. Where the eyedroppers really shine is when you need to quickly correct a color cast in a color photo. We'll cover the use of the Levels eyedroppers as well as the auto correction options i n detail in Chapter 7.
Levels Channel:
( Gray
f :I
( (
Input Leve l s :
t •
[251
( ( 11.00 I
[228 J
[Ol
Load . . . Save ...
) ) ) )
(
Auto
)
(
Options . . .
)
1/1.)' 1/1
Output Leve l s : •
6K Cancel
0
f255l
� Preview
Figure 5-23. The default Auto Levels adjustment. The shadow and highlight sliders are moved in to clip the endpoints by a value of O. l O%.
G lobal Image Improvements 1 1 39
Curves
Auto Color Correction Options r- Target Colors & Clipping
_ Highlights: c::::::J Shadows:
Clip: Clip
1[1[J % 10.10 1%
( (
OK Cancel
) )
( None Channel: ( Gray
,:1 EI (
Preset:
� :l
tf
( ( ( (
I
Save a s defaults
OK Cancel Smooth Auto Options . . .
) ) ) ) )
IYI Preview
'I--,.---
Figure 5-24. Witll a grayscalefile, the only option f01' the Auto a(lj'ustment is how much dipping is applied to the shadows and the highlights.
I
Output:
Understanding Curves The Curves d ialog box is arguably the most powerful tonal correction tool in Photoshop, a veritable Swiss Army knife for shaping the tonal and contrast character of an i mage. Anyth i ng that can be done with Levels can also be done with Curves, but the latter provides much more control over speci fic areas of the tonal range s ince it enables you to adj ust 16 d i fferent points anywhere along the curve i nstead of only th ree predefined poi n ts (black point, mid tone, and white poi n t) with Levels.
c,
..
Input:
[7r.1T7]
0 Show Cli pping
G Curve Display Options Show Amount of:
e Light
(0-255)
Pigment/Ink "
�
Show: � Channel Overlays � Baseline � Hjsto9ram
� Intersection line
I�
i12J�I�lf�r�["ll.�!!I1
A n atom y of the c urve
The use of a curve to express the tonal range of an i mage as well as the changes that are made to that tonal range exists in many i mage-edi ting programs. As a tool for mapping the tonal response of film emulsions, the curve even predates d igital edi t i ng by many decades. Though we are using screen captures from Photoshop CS3, Adobe Camera Raw, and Photoshop Lightroom to i l l ustrate how to work with a tonal curve, it's i m portant to understand that the controls for mani pulating the curve may d i ffer depending on what program you are using, but the basic a natomy of the curve, in terms of how it affects the tonal range, i s essentially the same (Figure 5-2 5) . T h e h istogram t h a t you see i n the Levels dialog box i s d i fferent for every i m age, b u t a n unaltered tone curve i s always a d i agonal l i ne over a g r i d run n i ng from t h e lower left corner to the upper-right corner. Once you know how i t operates, however, it's n o t so mysterious. W h e n configured to represent the tonal range as levels of l ight (as opposed to percentage of i n k coverage), the black poin t is i n the lower left corner, the w hite poin t is in the upper-right corner, and the m i d tone (Zone \1) is in the exact middle of the graph. From a visual perspective this makes sense as more l ight occupies the upper part of the curve while less l ight is in the lower part. (Thi nk of a d immer switc h that you slide up and down to change l ight levels i n a room.)
1 40 I Chapter
5:
Tone and Contrast
..
Tone Curve
Parametric \ Polnt.
=F 1 1/1 0
Highlights
Lights Darks
Shadows
0
0 __
0
r0-
0
O
0
r0-
0 O
Figure 5-25. The Curves intelfacesfor Photoshop CS3, Photoshop Lightroom, and Adobe Camera Raw. The Cu rves d i alog box i n Photoshop CS3. I n Photoshop
CS3, d isplay options at the bottom of the d i alog let you change the size of the grid from 4 x 4 to l O x 1 0 (Figure 5-26). You can also switc h between viewing the curves as light or a percentage of ink coverage. We recommend using the former because the idea behind i t, namely that you are adding and subtracting l ight, i s essentially the same concept as worki ng with i ncreased or decreased exposure, either i n the camera o r i n a traditional dark room. Using the l ight option also d isplays the tonal values using the 0 255 tonal scale, while the i n k option d isplays percentage values, and we feel it is simpler to stick with a s ingle measurement system when working with i m ages. When using the i n k percentage mode, the layout of the tonal range on the curve is reversed: d ragging up darkens the i m age (using more i nk), while d ragging down l ightens i t (usi ng less i nk). -
::!. Channel (}werb.ys I5?J e seline � Histogram g I tersectlon Une
Freehand m ode
Auto curves � �r-- Auto options
I
I n put value (before adjustment) C l i p p i ng display G rid size
Shadow, n eutral & h i g h l ight eyedroppers
Figure 5-26. The Curves dialog in Photoshop CS3.
I n versions of Photoshop p r i o r to CS3, switch ing between l ight and ink percentage displays is acco m p l ished by c l i c k i ng i n the horizontal gray ram p under the grid. You can change the grid size by Option/Alt-cl i c k i n g i nside the grid.
Other display options new to Photoshop CS3 are a Show Clipping option, which d isplays the same clipping prev iew that you can access in Adobe Camera Raw or Levels by Option/Alt-cl icking the Exposure/White Poin t or Shadows/ Black Poin t sliders. There a re also options (all turned on by default) that show a colored line to reflect how the curves i n the different channels have been adj usted, a baseline that reflects the unchanged curve, a h istogram , a nd i n tersection l i nes that track the poi n t location in the grid as you adjust it. There is also a P resets menu a t the top of the dialog with several common corrections and effects available. C l icking the Presets Options button to the right of this menu enables you to save your own curve adj ustments as presets. Curve and Levels: Comparable moves
Raising the cu rve above the i nitial d iagonal baseline increases the light, brightening the image (more exposure), while lowering i t decreases the light, darkening the i mage (less exposure). When such a change is accompl ished by d ragging up or down on the center of the cu rve (the m idpoint), the effect is the same as mov i ng the mid tone or gamma slider to the left or right in Levels (Figure 5-27) . T h e corner poi nts that represent black (lower left) and white (upper right) are the same as the I n put and Output endpoint sliders i n Levels (Figure 5-28). Moving the black poin t horizontally along t h e bottom of the curve grid determines the black poi nt, darkening the i mage, as well as potentially clipping darker shadow tones. This i s the same as moving
Figure 5-2 7. Raising the curve increases the light (or exposure) and lowering it decreases the light.
G lobal Image I mprovements [ 1 41
the I n put shadow slider to the right in Levels. Moving it vertically u p the left side of the grid lightens the black tones and is the same as mov i ng the Output Levels slider to the right. Movin g the white poin t horizontally to the left along the top of the curve grid sets the white poi n t, brightening the i m age and potentially clipping h ighlight values. This is the same as moving the I nput Levels high l ight slider to the left. Movi ng the white poi n t vertically down the right side of the cu rve grid darkens the white poin t and i s comparable to moving the Output Levels highlight slider to the left (Figu re 5-29). I n Photoshop CS3, black-and-white poin t sliders at the bottom of the curve grid function the same as the I n put sl iders in Levels. Channel:
! Gray
:1 ------,
Output Levels:
[Z] / [Zl o Show Clipping G CUI'Vf! Display Qptlons -----
ShowAmount ot Figure 5-28. The adjustments shown above in Curves and Levels would each have the same impact on the image: Shadow values of20 or below would be clipped, and highlight values above 230 would be clipped. Channel :
( Gray
:I
Figure 5-29. The adjustments shown above in Curves and Levels would each have the same impact on the image: Any blade tones of level 0 would be lightened to level 20, and any whites of255 would be darl(ened to 230. Finding tones and working with control points
To locate a specific i mage tone on the curve in Photoshop, move your cursor over that area of the image and hold down the mouse button. A small ci rcle will appear in the Curves grid, i ndicati ng the corresponding location on the curve (Figure 5-30) . To place a control point there, Command/ Control-click in the image. You can also place control points simply by clicking the curve itself. You can remove a poi nt by clicking it a n d d ragging it out of the grid.
1 42 I Chapter 5 : Tone and Contrast
S ��::t:�;I�� � [§[lEI Show: t. Channel Oved... � Biutllne � HiStogram I!IlnttrUCllon Un!! ys
Figure 5-30. Click a tone in an image to find the corresponding location on the tone curve. Command/Control-elide in the image to place a control point there.
I n Adobe Camera Raw, the way to locate an image tone on the c u rve is s l i g htly d i fferent: hold down the Com mand/Control key and m ove your m o u se over the i m age but don't cl ick with the m o u se button u n less you want to p l ace a contro l point on the cu rve at that location. In Adobe Photoshop Lightroom, s i mply move you r mouse over the image and the correspond i n g location will be s hown on the tone c u rve.
Control poi nts can be a d justed by d ragging them with the mouse, by using the arrow keys on the keyboard, or by entering values i n the Input and Output numeric fields. When using the arrow keys, a point will be adj usted i n single-level i ncrements. I f you a d d the Shift key to the a rrow keys, a point will be moved in l a-level i ncrements. The up/ down arrows change the Output (adj usted) val ues, and the left/right arrows change the I nput (before) values. Pressing the Tab key will switch between the I nput a n d Output fields, and you can select d i fferent control poi n ts by pressi ng
Control-Tab to move up the curve, or Control-Shift Tab to move down the curve. I n the Curves dialog i n Photoshop gray ramps appear at the bottom and on the left side. These are visual indicators of the input or "before" tonal values and the output or "after" tonal values. If you place a point on the curve and d raw an imagin a ry line down to the bottom of the grid, the tonal value for that point corresponds to the brightness level in that section of the gray ramp (input). When the point is adj usted, d raw an imaginary line from the poi n t towards the left side of the grid and the adj usted values will match the level of gray on the vertical gray ramp (output, Figure 5-3 1 ).
Channel: !,-� __ ra:.... y
.c-: 1 -----1
_ _ --,-
tutpur 130
The basic S - c u rve. One of the most useful adj ustments that can add a n essential snap to any i mage i s the simple S-curve. To apply a n S-curve, do the following: 1.
Click in the center of the grid to place a control poi nt. This will serve as the center pivot point for the S curve.
2.
Click in the middle of the top half of the cu rve and d rag up slightly. This raises the value of the highl ights, making them lighter.
3.
Click i n the middle of the lower half of the curve and d rag down slightly. This lowers the value of the shadow tones, making them darker.
The shape of the S-curve, how m uc h the highlights are raised and the shadows lowered, determines how much the contrast is increased. As you play a round with this, it should be clear after very l ittle experimentation that even a fai rly subtle S-shape leads to a noticeable boost in contrast (Figure 5-32) . The more exaggerated the shape of the S, the more d ramatic and pronounced the i ncrease in contrast will be (Figure 5-33). The objective when working with curves should be to keep the transitions i n the curve as smooth and fluid as possible and to avoid using too many control points. The more control points you place on the curve, the more unwieldy it will be to m a nipulate.
I t i s not necessary to use the center control point to make an S-cu rve adj u stment, but it does provide a handy pivot p o i nt that makes t h i s type of adj ustment a l ittl e easier when you are first gett i n g the hang of wor k i n g with a c u rve.
A reverse S - c u rve. If you are sta rting out with an image
Before
After
Figure 5-31. The gray ramps below and to the left ofthe curve grid show the level ofbrightnessfor the Input (before adjustment) Clnd Output (after adjustment) values. Control l i n g contrast
that you feel has too m uc h contrast, then a reverse S-curve can help to tame the contrast by reducing the d i fferences between the tonal extremes (Figure 5-34). To apply a reverse S-curve, follow the same steps as with a normal S-curve except reverse the positions of the upper and lower halves of the curve. The lower half will be raised up slightly and the upper half will be lowered down a bit. I n some cases you may find that one part of the tonal range will need to be adj usted more than the other.
As we pointed out earlier 111 this chapter, contrast i n a photograph is the difference between tonal values. To increase the contrast you make the lighter tones l ighter and the darker tones darker. Curves i s the ideal place to do this because it offers you the greatest control over how much the contrast is i ncreased and also which areas of the tonal range are affected most.
G loba l Image Improvements 1 1 43
•
Figure 5-32. In this image ofa desert canyon taken on an overcast day, a relativelyflat original is improved considerably by a gentle S-curve.
Mode->RGB Color.
Toni n g and Split Ton in g Effects 1 1 59
1.
Click the Add Adj ustment layer button at the bottom of the Layers palette, and then choose Hue/Saturation.
2.
In the H ue/Saturation d ialog box, click the Colorize checkbox i n the lower right. The i nitial hue will be based on the foreground color or if the foreground color was black, w hite or gray, it will be red. Move the Hue sl ider to change the color of the toning. A sepia or brown tone can be found between 25 (reddish brown) and 42 (yellowish brown); anything above 42 is likely to start looking a bit green.
3.
The default Saturation value when Colorize is checked is 25. I f this seem s a bit too saturated, move the slide to a lower number. For a very subtle effect that is barely noticeable as a sepia tone but evokes the look of warm-toned photographic papers, try a value between 6 and 1 2 (Figure 5-66 and Figure 5-67) .
4.
The Lightness adj ustment is not useful since it alters the luminosity of the original image i n a fairly heavy handed way. If your aim is simply to add a color tint to the existing tonal qualities of the image, then you should leave this adjustment alone. I t can be useful, however, if you want to d rastically darken all tonal regions of the image to mimic the look of dimly lit night conditions or antique processes from the 1 9th century, which often yielded prints of a dark brownish color. Hue/Saturation Edit:
Master
.
� I
Lightness: .
( ( (
.
I
Saturation:
.-f)
(
El
Hue:
..-G
.
G
[0
OK Cancel Load ... Save ...
) ) ) )
1
I J'I ;(
(�
e)
Coloriz
!J � Preview
•
Figure 5-66. The HueiSatumtion dialog box, with values to create a subtle Jepia tone.
Figure 5-67. Both images have been toned with a Hue/Saturation layer using a hue value of 35. The top image shows the effects of the default satumtion value of 25, which is often too strong. The saturation value for the bottom image has been lowered to 12 to create a more subtle warm tone effect. 0 Sf) Greg Gorman's warm-tone process
Noted portrait photographer Greg Gorman uses a n interesting method t o create his warm-toned black-and white images by i mplementing a mask created from a negative version of the i mage. The first part of the process incl udes an alternate way of converting to black and white that is based on the Lightness channel in Lab mode. While many people fa vor this technique for converting a color image to black and white, we feel that the new conversion features in Photoshop Lightroom, Camera Raw, and Photoshop CS3 offer more flexibility and interpretive control. We'll incl ude the Lab Lightness channel technique, however, j ust to give you some additional ideas and processing techniq ues for you r digital darkroom .
"'El ch05_steps.jpg
1 60 I Chapter 5: Tone and Contrast
Use the lightness C h annel for a black-and-white
4.
In the New Layer d i alog (Figure 5-69), open the Mode menu and choose Multiply. Click OK.
5.
I n the Color Picker, use the vertical spectrum bar and the l a rge Saturation/Brightness box to choose a color for the ton ing, and then click OK. The Multiply blending mode m ultiplies the values of the blend color with the base color (in this case, the i mage) and yields a result that is darker than the original image, so you will probably need to adj ust the layer's opacity slider to fine-tune the e ffect.
conversion
1.
Starting with a color photo, select I mage�Mode� Lab Color. In the Channels palette, click the Lightness chan nel to make it active.
2.
Convert to Grayscale by c hoosing I m age�Mode� Grayscale.
That is the basic method for using the Lightness channel i n Lab color to convert t o black a n d white (Figure 5-68) . The rest of the tech nique uses a negative version of the image luminosity to create a mask to control how the color ton ing is applied.
� Channels x (J'ailiSl
New Layer
-x
�
Lab
,...._..."..
See Figure 5-70 for a v iew of the layer mask. Figure 5-71 shows the result o f applying a Color Fill layer.
( OK ) Use Previous Layer to Create Clipping Mask ( Cancel) Color: ! DNone MOde: { Multiply :l pacity: � %
(
Lightness
)o
Pick a solid color:
t
�2
b
e •
�3
0H: lE:Jo OS: 189 1% OB Ii§ J%
l!QJiI!ilI'i Figure 5-68. After converting an RGB file to Lab COlO1; the Lightness channel is selected in the Channels palette prior to converting to Grayscale mode. load luminosity s e l e ction, Inverse, and Add C o l o r Fil l layer
1.
Cmd/Ctrl-click the G ray channel icon to load a selection of the brightness values in the image. Lighter tonal areas are more selected and darker areas are less selected.
o Only Web Colors
oR:1ill c§:] OB[loJ # lSi>3cOa I
OK
)
Cancel ) ( ( Add To Swatches) Color libraries ) OL:� Oa: III I Ob:� C:§:]% M:�% Y: 1100 1% K: Iso 1%
Figure 5-69. The New Layer dialog with the mode set to Multiply, and the Color Picker for choosing the color of the Fill Layer.
If you are u s ing t h i s technique on an image that is al ready in RGB mode and that has had a grayscale conve rsion applied u s i ng another method, you can load a selection of the l u minosity by Cmd/Ct rl c l i c k ing the RGB channel i n the Channe l s pal ette.
2.
Inverse the selection by choosing SeIect�Inverse, and then convert to RGB by selecting I mage�Mode�RGB Color.
3.
With the selection still active, add a Solid Color Fill layer by choosing Layer�New Fill Layer�Solid Color (if you use the Add Adjustment Layer button at the bottom of the Layers palette, be sure to hold down Option or Alt before you click to prompt the New Layer dialog to appear).
Figw'e 5-70. This is the layer mask that wa>' generatedfrom the inven-ed j'election of the luminosity (brightness values) of the grayscale image. To see the mask in the main document window, OptionlAlt click the thumbnail of the layer mask. OptionIAlt-clicl( it again to return to the normal image view.
Ton ing and Spl it Ton i ng Effects 1 1 61
The elegance of this technique is its use o f a layer mask based on the actual tonal values in the i mage to apply the color tone at varying strengths. Used this way, it could also be considered a split-toning technique since it applies the tone differently th roughout the i m age and blends it with the neutral gray values of the black-and-white conversion. When using a layer mask to apply a tone such as with this technique, remember that a mask can be edited a fter the fact if you wish to customize how the toni ng is applied. We'll cover more maski ng tec h n iques for toning in the following section on split toning, as well as in C hapter 6.
What About the Sepia Photo Filter?
Figure 5-71. 011 the bottom is the result of applying a Color Fill layer set to Multiply with a layer mask based on the il1versed luminance of the grayscale version. The initial effect was a bit too dark so the layer opacity was lowered to 70%.
The Photo Filter feature, w h ic h is available as a n adjustment layer, offers a Sepia filter a s well as n umerous other toning choices and the ability to choose a custom color. While this feature may seem like a natural choice and is certainly a quick and easy way to apply a sepia tone, we have never been that thrilled with the resu l ts. If you do want to try it out, keep in m i n d that the default Density setting of 25 produces a barely noticeable result for the Sepia fi l ter, and you'll probably need to increase this value significantly i n order to see a stronger sepia tone. Clicking the Color swatch lets you choose a new color from the Color Picker. For most i mages you'll want to leave Preserve Luminosity checked or the image will take on the brightness value of the color you have chosen.
Split Toning
Figure 5-72. To enhance the deep, rich tones and boost the contrast a bit, we used the Medium Contrast preset in the Curves dialog. €) Sf)
1 62 I Chapter 5: Tone and Contrast
Spli t toning involves using the d i fference between tonal values i n a n i mage, typical ly h ighlights and shadows, to apply the toning at different strengths, or to apply two d i fferent tones to the d i fferent regions. In the classic application of split toning, this would typically i n volve a cold or neutral tone a nd a warm tone. I n the traditional chemical darkroom, a split-toned effect could be created in a n umber of ways. One of the easiest and most common was to immerse a print in a bath of sepia toner for much shorter than the usual time. The result was that the highlight a reas of the print, which contai ned less silver, would be toned first, and by i n terrupting the normal d u ration of the sepia bath, the middle values and shadows were left i n the normal, neutral state. A further toni ng bath in a dilution of selenium toner would then shift the middle and neutral tones to a cooler tone or even a subtle violet shade i f the selenium d i l ution were strong enough.
Thankfully, achieving a spli t tone in the digital darkroom is a much easier and more controlled process, not to mention one that doesn't produce smelly chemicals! In this section we'll show you a few methods for directing the toning to different tonal regions as well as ways you can apply more irregular split or multi-toning that is not based on tonal values. Split toning in Photoshop Lightroom and Camera Raw
Both Lightroom and Camera Raw offer elegant methods to apply a split tone to images that are being treated as grayscale. The controls are the same for each program and consist of a set of sliders that enable you to choose a h ue (tone color) and a saturation level for the shadows and highlights, as well as a slider for setting the balance between the two. Although the split-toning feature in either program can be used with color images to create color effects, we'll be focusing on using it with grayscale images i n this section.
1.
Use the G rayscale Mixer controls i n either Lightroom or Camera Raw to create a grayscale conversion for your color image (see earlier i n this chapter).
2.
In Lightroom, the split tone controls are located j ust below the G rayscale Mixer (Figure 5-73) . I n Camera Raw, the spl i t tone controls are the same a s i n Lightroom and are found i n t h e tabbed panel j ust to the right of the HSLlGrayscale tab (Figure 5-74) .
3.
Choose a hue for t h e h ighlights a n d shadows, and then set the saturation (color intensity) level for each. If you want to create an e ffect that blends a color tint with the neutral gray areas of the black-and-white image, leave the hue for one of the tonal ranges (highlights or shadows) set to O.
4.
The Balance slider enables you to weight the toning more towards one area or the other. For example, choosing a negative balance setting will cause the toning color for the shadows to be more dominant throughout the image, and only the brighter areas will show traces of the highlight color. Choosing a positive balance value gives priority to the color chosen for the highlights, and only the darker values in the image will show the color that has been selected for the shadows.
If you want to apply a monochromat i c tone made u p of a single h u e in eit h e r Lightroom o r Camera Raw, set the hue val u es to the same color for both high lights and sh adows. T h e satu ration val u es can
The split toni ng controls in Camera Raw and Lightroom calculate the d i fferent tonal regions of h ighlights and shadows for you, which is one of the reasons they are so easy to use. Photoshop provides several ways to isolate the different tonal areas, however, and it's pretty easy to create your own masks based on i n formation already present in the i mage (Figure 5-75) . This first techn iq ue will show you how to use the Color Range command to create precise masks for the h ighlights and shadows that can be used i n adj ustment layers.
also b e set the same, o r you can expe riment with different satu ration values for the different tonal regions.
Toni n g and Spl it Ton i n g Effects 1 1 63
Color Range Select:
� DHighlights
:I CJ
( (
OK Cancel
(
Load ...
\
Save ...
) ) ) )
1"1£1£1 o Invert
o Selection Selection Preview:
0 Image
( None
:I
Figure 5-76. The Color Range dialog enables you to create a selection of sampled colors 01' jpecific tonal regions, such as the HIghlights that are being selected here.
Figure 5-75. Glinda's House. The anginal image and the version with sepia toning applied only to the highlights. ([) Sf)
'1l ch_05-gl i n da s h o u se.jpg 1.
From the menu, choose Select ..... Color Range. In the Color Range dialog, open the Select menu at the top, and choose H ighlights. Click OK to create a selection of the highlight tones (Figure 5-76) .
2.
Add a H ue/Saturation adj ustment layer (Layer..... New Ad j ustment Layer..... Hue/Saturation). In the Hue/ Saturation dialog, click the Colorize checkbox i n the lower right, and then adj ust the H ue slider to get the tone you want. I n the photograph "Glinda's House," we used a H ue setting of 35 and a Saturation value of 35 to c reate a warm, brown tone (Figure 5-77) . Since only the h ighlights were selected prior to adding this adj ustment layer, a layer mask has been created that shows only the toni ng i n the selected a reas (Figure 578). Click OK.
Figul'e 5-77. With a selection of the highlights active, add a Hue/ Saturation layer to create a jOepia tone for the highlights.
Figure 5-78. Applying the sepia tone to the l7Ighbghts USi17g a layer 117(lJ"/{ on a Hue/Saturation layer creates a split tone of the sepia and neutral gray tones of the original black-and-white photo.
1 64 I Chapter 5: Tone and Contrast
If you want to have a spl it tone that is created from a si ngle color tint, such as sepia, combi ned with the neutral gray tones of the black-and-white i mage, then you're done. 3.
To add a d i fferent color tone to the shadows or mid tones, repeat the above steps but select a different range of tones i n the Color Range dialog, and apply a d i fferent tint in the H ue/Saturation dialog.
4.
If you want to apply the second color tint to both the midtones and the shadows, Cmd/Ctrl-click the layer mask for the h ighlight tones to load it as a selection.
5.
Select-In verse to i n verse the selection and add a nevv Hue/Saturation adj ustment layer. For the Glinda image, we decided to go with a tint that is fashionable i n the Emerald City and used a h ue of 1 1 5 and a saturation of 1 0 to add a m uted greenish tone to the midtones and shadows (Figure 5-79).
6.
As a final step, zoom i n to 1 00% (View-Actual Pixels) and inspect the mask to see if there a re any areas that need touching up. To soften any hard edges on the mask c reated by the Color Ra nge selection, choose Filter-Blu r-Gaussian Blur from the menu, and apply a very slight blur of 1 to 1 .5 pixels.
Figure 5-80. The Way Down. The gmyscale version 01'1 the top and the f final version showing ilTegular application of two diferent strengths of sepia toning combined with the neutml gmyscale values. ,I) '
1.
A d d a H ue/Saturation a d justment layer and, in the dialog box, click the Colorize checkbox in the lower right. Adj ust the hue to 3 5 and the saturation to 12 to create a foundation sepia tone that is not too strong. Click OK. Double-click in the name for this layer and rename it "Base Sepia" (Figure 5-81 ).
Figure 5-79. Tilefinal split-tone ef f ect combines sepia for the highlights and a muted greenish tone for the shadows and midtones. Irregular toning by painting in layer masks
The masking method in the previous technique works well when you want to apply any toning o r tinting to specific tonal a reas. If you want a more uneven, i r regular application of the tone, however, you can use a si mple brush technique on the layer mask for the toning layer. This works well for combi n i ng both a color tone with the neutral gray tones of the original black-and-white image, o r for combi ning d i fferent strengths or "flavors" of a similar tone (Figu re 5-80).
Figure 5-81. The base Sepia layer adds a foundation tone to the image that is noticeable but not overpowering.
Toning and Split Toning Effects 1 1 65
2.
Make a copy of the first H ue/Saturation layer by choosing Layer --->New--->Layer Via Copy (Command or Control-J). Double-click the thumbnail for this layer to open the H ue/Saturation dialog. Using the same h ue value, i ncrease the saturation to 2 5 or 30 to create a much stronger application of the brown tone. Click OK.
3.
Double-click i n the layer name for the second H ue/ Saturation l ayer and rename it "Strong Sepia." I nvert the layer mask for this layer by choosing I m age ---> Adjustments---> I nvert (Command- or Control-I). This will turn the layer mask black and will h ide the strong sepia effect (F igure 5-82) .
Figure 5-83. Painting with white at varying opacities on the black layer mask causes the stronger tone to show through in an irregular fashion.
Figure 5-82. Inverting the layer masl( t o black hides the strong sepia effect.
4.
Select the B rush tool. Click i n the Options Bar to show the B ru s h Picker, and select a large, soft-edged brush. The brush size you use will vary depending on the size of the i mage you a re working on, but a brush size of 300 to 500 pixels is often a good size to start with. Set the brush opacity in the Options Bar to 25%.
5.
Press "D" on the keyboard to senhe foreground color to white (when a mask is active, white is the default foreground color, but when a layer is active, black is the default). In a n uneven a nd i rregular fashion, begin to brush in the stronger "flavor" of sepia in d i fferent parts of the image. The idea here is not to tone specific elements i n the image, but to add the ton in g i n a seemingly random m a nner. Vary the size of the brush as needed (Figure 5-83).
6.
Repeated brushstrokes over areas you have already pain ted on will cause the gray values in the layer mask to become lighter, which will i n turn cause the tone to become stronger as more of it is revealed. Press "X" on the keyboard to exchange the foreground and background colors and paint with black to scale back the appearance of the stronger sepia tone.
7.
Painting with black using a low brush opacity on the Base Sepia layer will dilute that tone and cause traces of the original neutral grayscale tones to show through (Figure 5-84) .
1 66 I Chapter 5 : Tone and Contrast
Figw'e 5-84. Thefinal image .) S/J
Combining Color with Black-and-White Like toning, accenting black-and-white photographs with color has been around for a long time. For many years the primary way this was achieved was by painting on photographic prints. Combin i ng color with black and white can result i n i n teresting creative treatments that enable you to focus the v iewer's attention through visual signals created by the selective use of color. This effect can be achieved using the origin a l i m age colors of a digital capture or a color scan, or by selectively adding color to a black-and-white image. In this final section of the c hapter we'll cover using both of these approaches with your photographs.
Blending original image colors with black-and-white
Using original colors as accents in a black-and-white version of an image is really more about selection and masking techniques. No matter what technique you use, accurate selections and precise masks are requ ired to reveal specific areas in color. Of course, you may choose to use a more painterly approach w here exact edge fidelity is not that critical, such as i n the previous example on i r regular sepia toning, but however you do it, the use of layer masks will be required. The selections or masks that you use for this technique can be created either before you apply the black and-white conversion or after (Figure 5-85).
3.
Set the Tolerance to 40 (the default), and click in the orange of the l i fe preserver. Click the "plus" eyedropper (or hold down the Shift key), and click i n areas o f orange until the preview shows the l i fe preserver as white (Figure 5-87). Click OK to create the selection (F igure 5-88).
Figure 5-86. Selecting an area to target with Color Range will enlarge that area in the preview. Color Range Select:
I
Color Range
:1 � f40l � � � �
" Sampled Colors
Fuzziness:
Select:
!
:1 � f40l � � � �
" Sampled Colors
Fuzziness:
olnven
Selection Preview:
! None
:1
Olnven
SelectIon Preview:
I None
:1
Figure 5-87. On the left is the initial selection previewed as a mask. On the right is how the selection looks after adding areas of orange that had not been selected by thefirst sample click·
Flgure 5-85. By making only the life preserver in colOl; attention is focused onto that area of the image.
Ii) S1)
"'EJ ch5Jife_prese rver.jpg 1.
Open the ch5 Lije...jJreserver i m age and add a Black & White adj ustment layer. Create a grayscale conversion, and then click OK.
2.
Click the view icon for the black & white layer so it is not visible, and then click the Background layer to make it active. With the Rectangular Marquee, drag a selection marquee around the life preserver (Figure 5-86). From the menu, choose Select�Color Range. By selecting a rectangle around just the l i fe preserver before bringing up the Color Range dialog, it will preview only that area of the i mage and make it much easier to see what we a re doing.
FIgu re 5-88. The selection created by Color Range.
Toning and Sp l i t Ton i ng Effects 1 1 67
T h e Qu ick S e l ection Tool also works great for areas of d i stinct color s u c h as the orange of the l ife preserver.
4.
5.
Make the Black & White adj ustment layer v isible. Press "D" on the keyboard to make black the background color, and then press the Delete key to fill the selection with black. This masks the grayscale effect from the orange life preserver (Figure 5-89). Zoom in to 100% (View--->Actual Pixels) a nd inspect the mask to see if there are any areas that need touching up. To soften any h a rd edges on the mask created by the Color Range selection, choose Filter---> Blur--->Gaussian Blur from the menu, and apply a very slight blur of 1 to 1.5 pixels.
In this fi nal section of the cha pter we'll cover a few of the methods for creating a hand-colored look . As with our overall philosophy about working in the digital d arkroom, the techniques that we'll show are designed to be flexible and non-destructive. This means that the application of color will be done with layers, adjustment layers, and layer masks. Tinting with adjustment layers and layer masks. One
of the easiest ways to add selective tinting is to use a series of Hue/Saturation adj ustment layers with layer masks that target the color tint to specific areas of the image. The qual i ty of your selections and masks will determine the qual ity of the tinting, so taking the time to make good selections and masks is important. If your i mage has only a few areas that need tinti ng, and the s hapes are not too i n tricate, then completing the image will be relatively quick and easy. The more complex the i mage and the coloring requ i rements, the more time it will take you to finish the task. In order to concentrate on concept and not get too bogged down by a lot of detail work, the example we'll show here (Figure 5-90) has some fai rly simple tinting requirements.
Figure 5-89. The mClsk protects the life preserverfrom the effect rendered by the Blacl( & White adjustment faye/: Color tinting
Color tinting i nvol ves adding color selectively to a black and-white i mage (or a color i mage with a grayscale effect) to create what is commonly known as the "hand-tinted" look. In the days before the wonders of the d igital darkroom, this effect was accomplished by applying special photo oil paints directly to the surface of a print. I t was time consuming, smelly, and the paint took q uite a while to d ry. Though adding color tints digitally is m uch easier and can be done with more control than by hand, for some artists the key factor about applying the colors by hand is the actual involvement of their hands. Many photographers who have done a lot of traditional hand coloring still miss that human touch in the a rt making process. Sean still has some of his photo oils, though the caps are firmly cemented in place d ue to yea rs of neglect.
1 68 I Chapter 5: Tone and Contrast
Figure 5-90. The pumafountain.
.
SI)
'1l ch5_puma-fountain.jpg
1.
2.
3.
Open the file "puma-founta i n . j pg." This has already been converted to grayscale. To add color tinti ng, choose I mage->Mode->RGB Color.
( (
Use the Lasso tool with a feather val ue of 0 to make a selection of the center brick work surrounded by the carved puma head. Total accuracy is not that critical. Hold down Option/Alt to subtract from the selection, and d raw around the center square j ust behind the puma head. With the selection active, click the Add Adj ustment Layer button at the bottom of the Layers palette, a nd add a Hue/Saturation layer. I n the H ue/Saturation dialog box (Figure 5-9 1 ), click the Colorize checkbox and move the hue slider until you get a reddish brick color. We found that a hue setting of25 and a saturation of20 worked pretty well. Click OK.
4.
Gaussian Blur
The mask edges a re too hard, so we'll soften them with a blur (Figure 5-92) . From the menu, choose Filter->Blu r->Gaussian Blur. Enter a bl u r radius of about 8 or 1 0 pixels, and then click OK. This soften s t h e edge j ust enough so the e y e can't detect a hard transition.
Radius:
[25J
) )
100%
EI
ll�O_�J
pixels
.------{Or---� Figure 5-92. Adding a slight blur to the hard edges of a masl( disguises the transitionsfrom one color to anotheJ: 5.
Now make a loose selection of the puma's head with the Lasso tool. Add a new H ue/Saturation adj ustment layer (Figure 5-93) . I n the dialog box, click the Colorize checkbox and move the hue slider until you get a tawny tan color that would work for a mountain lion (of course, you're also welcome to make a punk lion if the spirit moves you! ) . We used a h ue o f 42 and a saturation of 32. Click OK.
6.
Press Command/Control-F to apply the same Gaussian Bur setting to the new layer mask and soften its edges.
7.
Add a new H ue/Satu ration adj ustment layer (Figure 5-94) . Click Colorize and move the hue slider to about 2 1 5 to create a bluish color that would work well for the wall. Set the saturation to 1 5, a n d then click OK. Don't be concerned that this layer covers all your previous ti n ting. We'll fix that i n the next step.
8.
Press D on the keyboard to set the default foreground/ background colors, and then press X to exchange them and place black in the foreground swatch. I n the Options Bar, choose a brush that i s o f a good size to cover up the center brickwork, and set the brush opacity to 1 00%. Paint over the center bricks to mask that part of the blue tint and reveal the brick color from u nderneath. Lower the opacity of the layer to 65%.
Muttr --:l Hut:
Cancel
� Preview
Hut/Satur-uion £dl\
OK
Figure 5-91. Thefirst Hue/Saturation adjustment layer adds a tint to the brickwork.
Ton i ng and Split Ton i ng Effects 1 1 69
to work, the blending mode for the layer needs to be set to Color. While the adj ustment layer and masking method we have used so far with this example offers the greatest flexibility, i t entails that each item that is a d i fferent color has to have its own separate adj ustment layer and mask. By pai nting on a n em pty layer, you mix colors together as an artist may do on the canvas, making i t ideal for adding small accents of different colors. Since very few thi ngs i n l i fe a re · all one color, the addition of other h ues helps to break up the continuous tint of a single area. 9.
A d d a n e w layer t o t h e top of t h e layer stack (Layer-.. New Layer). Set the blending mode for this layer to Color.
10. With smaller, soft-edged brushes, begin add i ng irregular splashes of tan a nd darker browns to the blue-gray areas of the wall. If a dab of color seems too bright or i n tense, switch to the Eraser tool (E), choose an opacity of 25%, and begin to gradually erase it until the saturation and density of the color looks OK.
1 1 . Switch back to the B rush tool and choose a drab Figure 5-93. The second adjustment layer adds a tan tint to the pumas head.
greenish color from the Swatches palette, or click i n t h e Foreground swatch a nd choose a color i n the Color Picker. Paint at lower opacities on the area j ust under the puma's mount, which is where the water spills from when this fountain is turned on. The goal is to add just a slight h i nt of green color here to suggest the idea of moss growing on the rocks (Figure 5-95).
Figure 5 -94. The thi,·d adjustment layer tints the stones of the wall with a blue-gray colo!: Tinting by painting on an empty layer. Another way to
add color tints to an i mage, though not q uite as flexible as usi ng adj ustment layers and layer m asks, is to paint d i rectly on an empty layer that sits above the main image and any other tinting layers you have added (Figure 5-95). For this
1 70 I Chapter 5: Tone and Contrast
FigUl'e 5-95. The final tinting is painted directly on al1 empty layer that uses the Color blending mode. 0 Sf)
Figure 5-96. "Chips & Salsa with Frida KaMa " is a 45-second exposure photographed with a homemade pinhole camera on black-and-white negativefilm. The tonal correction and color tinting was accomplished using neady 30 adjustment layers with layer masl(s. Blur-->Gaussian Blur and use a radius of20 pixels to soften the edge of the mask.
Figure 6-36. After darkening select paw of the image. Adding emphasis through selective burning
The first step is to darken everyth i n g in the image except for the sign. Since the sign is an i m portant element i n the scene, we want it to stand out from the background. Darkening the background is one way to accomplish this, as well as lowering the tone of the i mage to create a darker, more foreboding mood.
'1:J ch06_N oVi s itors.jpg 1.
Use the Polygonal Lasso to make a selection around the stop sign and the "no visitors" sign (Figure 6-37) . Don't worry about exactly tracing t h e edges, s i n c e a blur will be applied to the layer mask we create from this selection. Choose Select-->Inverse to change the selected areas from the sign to everything else except the sign.
Non-Destructive Dodging and Burn i ng 1 1 87
Figure 6-40. The mugh selection of the sky. 6.
Acid a Curves adj ustment layer. Place a poin t on the upper part of the curve (approximate r nput value of 2 1 0) and d rag down to signi ficantly darken the clouds (our Output value was 1 50). Click O K (Figure 6-4 1 ).
7.
Apply a Gaussian B l u r to the layer mask for the da rken sky curves layer. A Radius o f 2 0 to 25 should work well to soften and disguise the hard edge. Click OK to blur the mask (Figure 6-42) .
Figw·e 6-38. CmdlCtrl-dick to place control points on the curve.
Figure 6-39. Adjusting the curve to darken the landscape and preserve the highlights on the road. 5.
Now it's time to darken the sky. Cmd/Ctrl-click the thumbnail of the layer mask to load a selection based on the white areas of the mask. Select the Rectangular Marquee tool and Option/Alt-drag over the bottom half of the image to subtract the land from the selection (Figure 6-40) . This does not need to be a totally accurate edge along the horizon since this will eventually be blurred i n the next layer mask.
1 88 I Chapter 6: Dodging, B urn i ng, and Exposure Control
� 12 10 I 0/1ZJ OShow Clipping
Figure 6-41. Darl{ening the sky.
Ra.dlus: [20] pixels e
Figure 6-42. Blurring the Darl(en Sky layer m ask.
8.
9.
The d i rt road is a nother key element in this i mage. By lightening i t slightly, it becomes more promi ne n t. Use the Polygonal Lasso tool to trace a rough selection of the road (Figure 6-43 ) . Add a Cu rves adj ustment layer and click OK w ithout making any cha nges.
Figure 6-44. The lightened road.
Set the blending mode for this layer to Screen. Lower the Layer opacity to 60% . Soften the edge of the road layer mask by applying the same Gaussian B l u r of 20-25 pixels (Figure 6-4 4 ) .
The step-by-step instructions featured here will enable you to duplicate what we did to this i mage. Far more important than just following our steps is understanding the overall concepts of how to make these type of interpretive changes to your own photographs, and why you might choose to do so to enhance and interpret your own images.
As with the Death Valley i mage, we have been able to transform the character of this photograph through the use of i nterpretive tonal adj ustments. In the case of this image, most of those have darkened the scene.
B u rning in the Edges In the "No Visitors" i mage, significant d a rkeni ng was used to suggest a forbidding mood. But not all darkening adjustments create a d a rk or gloomy feeli ng. I n many images, darkening a reas is a nother way to sculpt and mold the tonal qualities in the i mage and give more visual emphasis to certain parts of the photograph. While the emphasis created by l ightening is obvious to u ndersta n d , you can also d raw the viewer's attention to key a reas by d a rkening the regions a round them, which will make the adj acent a reas seem lighter. A classic use for such a technique is with portraiture. Creating a n oval-shaped vignette or edge burn is an excellent way to darken the a rea a round the subject's face and make the face the focal poin t of the i mage (Figure 6-45) .
Figure 6-43. The selection of the dirt road.
Non-Destructive Dodging and Burn i ng 1 1 89
Figure 6-46. The curve to create the darkened vignette.
Figure 6-45. Before and after burning in the edges.
«) Sf)
'1l ch06_po rtrait.jpg 1.
Use the Elliptical Marquee tool to create a n oval selection around the subject's face.
2.
Choose Select--->I n verse to select everything except the face. Add a Curves adjustment layer and d rag down the center of the curve to darken the a reas surrounding the face (Figure 6-46). I t's OK to make the adj ustment a bit darker than you thi n k i s correct, since this can be modified later using the layer opacity controls.
3.
Remove the hard edge by applying the Gaussian Blur filter to the Curves layer mask . Blur to taste (we used a 60-pixel radius). The object of the blur is to disguise the edge so you cannot see where the adjustment begins or ends (Figure 6-47).
1 90 I Chapter 6 : Dodgi ng, B u rn i ng, and Exposure Control
Figure 6-47. A Gaussian Blur softens the edges of the layer mask.
4.
We decided to mod i fy the layer mask by painting with black at a brush opacity of 40% to remove some of the darkening e ffect from the lower strands of the girl's hair (Figure 6-48) .
Figure 6-48. The layer mask was modified to protect the girl's hair from the darkening effect.
Using a n oval vignette to burn i n the edges of a portrait is a technique that has been around for nearly as long as photographers have been making portraits. And i t works well on other types of images, too, so give it a try on some of you r own photographs. Irregular or painted edge burn
The previous example of burning in the edges works well for portrait subjects where you want the face to be the lightest element in the i m age. You can use the same basic technique with a rectangular marquee or an u neven lasso selection to create a burned-in e ffect that is applied only to the outer edges of a photo (Figure 6-49) .
Figure 6-49. The Ship House image before (top) and witl2 an irregular edge bum (bottom). ') 'in
JEJ ch06_S h i p H o u se.jpg 1.
Use the Lasso o r Polygonal lasso to create a loose, i rregular selection j ust i nside the edges of the i mage (Figure 6-50) .
2.
Choose Select->Inverse and add a Curves adj ustment layer. Click OK without m a k i ng any changes to the curve. At the top of the Layers palette, set the blending mode for the layer to M ultiply to darken the outer edges of the photo (Figure 6-5 1 ).
3.
Choose Filte r->Blu r->Gaussian B l u r a nd set the radius to a h igh n umber (we used 50) to apply a strong blur to the layer mask. As an optional customization step, you can use the Brush tool and a large, soft-edged brush to further modify the look of the edge burn (Figure 6-52) .
Figure 6-50. The irregular lasso j'electioll just inside the edges of the image.
Non-Destructive Dodging and B urning 1 1 91
C ontrast and Expos u re C ontr ol Contrast is a n essential ingredient i n photographs and o n e that needs t o b e present i n j ust t h e right amount. I f there is too much contrast, the highlights are i n danger of blowing out to total white and subtle shadow detail is lost. B u t i f there is not enough contrast, then the image is fl a t and dull. And the right amount of contrast is not something that is easily quantified as it can vary from i m age to i mage depend ing on the subject matter, the purpose of the photograph, a nd the personal preference of the photogra pher.
Blending Mode Contrast Enhancement
Figure 6-51. The effects of the empty Curves adjustment layer set to Multiply to burn in the edges.
Many times using Overlay, Soft Light, and Hard Light layer blending modes adds the needed contrast boost and color saturation. These modes work best on i mages that do not have a lot of existing contrast. If there is too m uch contrast i n the original file, then the effect may be too pronounced, though lowering the layer opacity can sometimes help with this. Empty adjustment layers and contrast-enhancing blending modes
In the exam pie of the tree and the m issing steps, the Overlay blending mode provides both a d ramatic contrast boost as well as a significant increase in color satu ration (Figure 6-53). Because the image was not overly contrasty to begin with, there are no major problems with using this blending mode.
Figure 6-52. Blurring the layer mask to soften the edge burn effect.
Figure 6-53. The Missing Steps, before and afta The modified version uses an empty a(y'ustment layer set to Overlay blending mode. « .1/ )
1 92 I Chapter 6 : Dodging, Burning, and Exposure Control
'1J ch06_M i ssingSteps.jpg 1.
Add a Curves or Levels adj ustment layer and click OK in the adj ustment dialog without making any change (Figure 6-54).
2.
Set the layer mode to Overlay (Figure 6-55). I f Overlay seems too strong, lower the layer opacity, or try the Soft Light blendi ng mode (Figure 6-56).
Figure 6-56, Changing the blending mode to Sofi Light provides a more subtle contrast enhancement tha n Overlay. Targeted adjustments w ith blending modes, the Blend If sliders and layer masks
Figure 6-54. An empty (meaning no changes were made) Curves adjustment layer set to Nonnal mode does not change the image in allY way.
Many times creating a tonal adjustment with blending modes may be well-suited for some parts of the tonal range but not others. The Blend If sliders found i n the Advanced Blending Options dialog are one way to d i rect the effects of a blending mode to only parts of the image (Figure 6-57).
Figure 6-55. An empty CU1'ves a djustment layer set to Overlay resultj' ill a j'ignijicant increase in both contrast and color mturation,
Figure 6-57. T-Rex, be/m'e and afie!: The Blend Ifsliders tCl1get only the brighter tones in the photo with a lightening adjustment , " 1" / )
Non-Destructive Dodging and Burn i ng 1 1 93
layer Style
'1J ch06_T-Rex.j pg A Ty ra nnosaurus Rex skeleton is always an i nteresting subject for a photo, but the quality of museum lighting for such d isplays va ries widely a nd some bones a re better lit than others. I n the photograph for this exam ple, although a spot light is d i rected on the skull, the ligh ter tonal val ues are still a bit dark and the separation between the skull and the background could be better.
1.
Add a Cu rves adjustment layer and click the OK button without making a n y changes to the curve. I n the Layers palette, set the blending mode for the adj ustment layer to Screen (Figure 6-58). This will create a noticeable l ighte n i ng of the entire image.
Styles
Blending Options: Custom OOrop Shadow
Qlnner Shadow
� aOuler Glow
I
Dinner Glow
08evei and Emboss Comour
OTtlllUrl!
- Blendlng Oplfons
I
Blend Mode: r Screen Opacity:
FIJI OpaCity:
Channels: � R
{!:Ie
a fiOOl"
�B
Knockout; �
� Blendlnterlor EffeCl5aS Group
Blend Clipped Layers as Group
I
� � ( New Style... ) � Prevlew
C
Transparencv Shapu Layer
QColor OV!rlay
o Layer Mask Hides Effem o VUlor Muk Hides Effects
I O Gradlenl overlav O Stroke
:1 9 fiOOl"
, Advanced Blending
GSatin
OPoittern Ovtrlav
;:::;:===:::;:;:;--,
.Genual Blendln •••
Bltnd lf:� This laver.
255
Undertylng Layer:
255
Figul'e 6-59. The Blend Ifsliders in the Layer Style dialog.
Figure 6-58. The initial lightening effect created by em empty Curves adjustment layer a11d the Screen ble11ding mode. 2.
Double-click the a rea to the right of the layer's name to bring up the Layer Style d ialog box. In the bottom section of this dialog are the Blend If sliders, which control which tonal values on the active layer are visible in the image (Figure 6-59). I n the case of the empty adj ustment layer used in this tech nique, it controls which tonal values in the image a re a ffected by the adj ustment.
3.
In the Blend I f section, move the This Layer shadow sl ider in to level 2 5 (Figure 6-60). This means that any tonal values o f 2 5 or below a re not a ffected. But this creates an abrupt, noticeable transition. To smooth the transition, split the slider by Option/ Alt-cl icking on the right half and moving it to the right until the split value on the right of the slash is 45 (Figure 6-6 1 ).
Figure 6-60. The initial adjustment of m oving the shadow sliderfor This Layer to 24 results in posterized tonal transitions in some areas.
Figure 6-61. OptionIAlt-clid( on the slider and drag to split it into two halves. Thisfeathers the effect and creates smoother tramitiollS between tones.
1 94 I Chapter
6:
Dodgi ng, B u rni ng, and Exposu re Control
The split between the numbers creates a transitional feathering effect so the blend is seamless. In this case, the adjustment layer is affecting no tonal values darker than level 25. As the values i ncrease above 25, they are gradually a ffected until level 45, when the adjustment is applied at full strength. The Blend If sliders allow the lighter areas of the image such as the highlights on the skull and on the wall, to be lightened, while protecting the darker areas from the effect. 4.
At the top of the Layer Styles dialog box, you can try lowering the opacity to fine tune the e ffect. For the version we created, we chose to lower the layer opacity to 80%. Click OK to apply the blending changes.
5.
Some areas on the skull or the background may still be too bright. Using a soft-edged brush tool, paint on the layer mask with black at 30% to gradually tone down the areas that need further darken i ng. And remember that you can also ad j ust the curve to refine the effect further. For our final version we l ightened the image a bit more by adjusting the curve (Figu re 6-62) . - Channel: IRCB
1.
I n the i mage of the DC-3 , use the Polygonal Lasso tool with a feather value of zero to make a selection of the black area where the w i ndows a re located.
!t
Figure 6-63. Before an d after: the adjustments are accomplished using blending modes, layer masks, and the Blend Ifsliders. 0 SIJ
Figure 6-62. Thefinal tweaks for this effect included lowering the layer opacity, modifying the layer masl( and adjusting the curve.
2.
Select the Magic Wand tool, set the Tolerance to 32 (the default), hold down the S h i ft key, and click i n the sky to add that a rea to the selection. Choose Select I nverse to reverse the selection so that only the light metal a reas of the fuselage are selected.
3.
Add a Curves adjustment layer and in the Curves d ialog click OK without making any changes. Set the blend i ng mode for this layer to Multi ply. This provides a n ice darkening e ffect that works well for the reflective metal (Figure 6-64).
4.
Double-click to the right of the layer name to open the Layer Style dialog box. In the Blending Options panel, go to the Blend If sliders and move the h ighlight sl ider
Modifying contrast w ith blending modes, the Blend
If sliders and layer masks Now let's explore another variation of this technique using a different blend mode and more precise layer masks. Our subject for this example i s a detail v i e w of an old DC-3 ai rplane from the 1 930s (Figure 6-63).
Non-Destructive Dodging and Burn ing 1 1 95
for This Layer to 1 50. Option/Alt-click the slider to separate it i nto two halves. Move the right half to the right until the values for this slider read 1 50/200 (Figure 6-65) . Click OK.
5.
Cmc\/Ctrl-cl ick the layer mask to load it as a selection. Choose Select--->I nverse. Select the Lasso tool, press Option/Alt, and d rag a selection a round the sky to deselect that a rea. The objective is to have a selection of only the black strip over the wi ndows.
6.
Add another Curves adj ustment layer and click OK i n the Curves dialog box without making any changes. Set the blending mode for this layer to Soft Light. You can also try Overlay but we liked how the windows looked with Soft Light (Figure 6-66).
7.
I f any of the layer mask edges look too hard , apply a slight Gaussian Blur to soften them.
Figure 6-64. The empty Curves layer set to Multiply darkens the bright highlights (a layer mask protects the dar/{er areas).
Figure 6-66. The final layersfor the blending mode contrast adjustments to the DC3 photo.
High Pass Contrast Enhancement
Figure 6-65. Left: Moving the This Layer highlight slider to 150 resullS il1 rough transitional edg6. Right: Splitting the slider and placing the two halves at 150 and 200 create a smooth transition between areas that are being adjusted and those that are not.
The split between the n umbers of the This Layer sliders means that no tonal values from the Cu rves Multiply laye r brighter than level 200 a re bei ng used. As the values progress towards level 1 50, more and more become visible. The effect of the Cu rves Multiply layer shows at full strength from level 1 50 and below.
1 96 I Chapter 6: Dodgi ng, Burning, and Exposu re Control
As we have al ready seen in numerous examples, contrast is an important element in shaping the tonal qualities of an image. While contrast that is applied to the brightest and darkest a reas of the image is perhaps the most noticeable, the contrast in the mid tone a reas of an image is especially crucial, and the effectiveness of any adjustments you make to this important area can have a great impact on the success of the image. In many cases, Curves is our preferred tonal correction tool when we need to apply adjustmen ts to specific regions of the tonal range, but as precise as i t is, it makes no distinction between textured a reas, smooth areas, or a reas that contain depth created by the contours of shape and form.
One way to achieve a contrast boost that affects only the contoured parts of an i mage is to take advantage of the High Pass filter, the Overlay blending mode, and the Blend I f sliders (Figure 6-67). We fi rst learned this great technique from master digital printmaker R. Mac Holbert. This is a modified version that takes advantage of the non-destructive flexibil ity of Smart Filters i n Photoshop CS3.
2.
On some i mages, the end result of this techniq ue can cause some slight color s h i fts along contrast edges. Since we only want this effect to apply to contrast and not to color val ues, we'll remove all the color saturation from the dupl icate layer. Choose I mage-> Ad j ustmen ts-> Desatu ra teo
3.
Convert the Layer i nto a Smart Object by choosing Filter->Convert for Smart Filters (Figure 6-68) .
Figure 6-68. The dej'aturated background copy layer converted to a Smart Object.
4.
Double-click the Background copy layer just to the right of the layer name to open the Layer Style dialog box. Set the Blend Mode to Overlay (Figure 6-69) and the Opacity to 75% (this is just a starting number that can be adjusted as needed once the effect is in place). In the Blend If section move the black slider for This Layer to 50. Hold down Option!Alt and d rag on the right half of the slider to separate it into two halves. Move the right half to 80. Now move the white slider to 1 75, hold down Option/Alt to split the slider, and move the right half to 205 (Figure 6-69). Click OK to apply the changes.
5.
Choose Filter->Other->H igh Pass and set the radius to 50 pixels. Click OK (Figure 6-70) .
6.
The Blend I f n umbers mean that no tone d a rker than 50 or brighter than 205 will be affected by the contrast boost from the H igh Pass layer. The spl it from 50 to 80 and 205 to 1 75 creates a transitional feathering effect, spa n n i ng 30 tonal levels where the contrast e ffect gradually becomes more visible. The end result is that the contrast increase is only visible i n the middle parts of the tonal range and the effect of the blend i ng mode and the H igh Pass filter enhances the existi ng contrast along the contours in the i m age
Figure 6-67. The original image (top) , and after the High Pa;-s Contrast technique (bottom). :u Sf)
"'el ch06_h i gh pass_co n t rast.j pg 1.
Choose Layer->Duplicate layer to make a copy of the Background layer. If you have a l ready applied any adj ustment layers, the H igh Pass e ffect needs to be applied to the cumulative result of all your layers. To create a copy layer based on all of the layers, make sure that the topmost layer is active and hold down the Option/Alt key and choose Merge Visible from the Layers palette pop-out menu. Keep the Option/Alt key down until you see the new layer thumbnail appear in the Layers palette.
Non-Destructive Dodging and B urning 1 1 97
7.
'Nith the basic effect i n place, you can now fine-tune the High Pass filter settings. In the Layers palette, double-click on the name of the H igh Pass Smart Filter. Experiment with d i fferent radius settings to see how the contrast is a ffected (Figure 6-71 ) . We decided on a radius setting of 8 for this example, but it's important to understand that there is no "right" value here that works for all images.
Advoln
I nverse.
2.
Add another Curves adj ustment layer and create a curve to l ighten and boost contrast in the lower half of the i mage (Figure 6-92) .
For this i mage of t h e Death Valley road, we also made a Polygonal Lasso selection of the road and added a Cu rves adjustment layer to further lighten and add contrast to the road (Figure 6-93) .
Contrast and Exposure Control I
207
Controlling Dynamic Range I n photography, dynamic range is the range of tonal values from the da rkest s hadows with detail to the brightest highl ights with detail . Cameras-film or d igital-cannot record the entire d ynamic range that is visible to the human eye, and it can be very challenging to deal with h igh-contrast scenes that contain a dynamic range that is greater than a camera can record in a si ngle expos u re. The cu rrent state of-the-art of digital sensors have a dynamic range that is very similar to slide fil m , which is roughly about six stops. In situations with h igh contrast, this limited dynamic range can mean sacrificing the detail at one end of the tonal scale to retain detail in the other. With digital capture, exposures are usually made to hold detail in the brightest highlights, which results i n s h adow detail that may be much darker than you would like. Fortunately, i f your images originate as raw digital captu res, there a re ways to control h igh contrast and even artificially expand the limited dynamic range of you r camera's image sensor. Multiple raw processing of a single exposure
Figure 6-92. An inverted version a/thefirst layer 111asl( is used to apply a curve that lightens and adds contrast to the lower part a/the photo.
The flexible nature of a raw file means that it can be easily interpreted in d i fferent ways. Applying diffe rent conversions to a raw file and combining them in Photos hop is a useful technique for extendi ng the tonal range of an image and it's one that does not requ i re a tripod or shooting m u ltiple exposures. In an image with contrast problems, a raw file can be processed more than once to opti mize the tonal range of specific a reas (Figure 6-94) . The fi rst processing, for example, might be an overall treatment that preserves detail in the bright h ighl ights and creates a good brightness level for the midtones ( for this to work you need to make sure the h ighl ights a re not clipped when the exposure is made). In the second process ing, the file could be significantly lightened to open up the darker s h adows and reveal i m portant detail in those a reas. The highlights might be blown out i n this second version, but only the lightened shadow a reas would be visible in the fi nal composite of the two images. Here are the basic steps for processing a single raw fil e twice and then combining the results into a single image with a wider contrast range.
Figure 6-93. Thefinal curves a(lj·ust111ent to lighten the road.
208
I Chapter
6:
Dodging, Burning, and Exposure Control
Figure 6-95. Setting up the Workflow Option s so the raw file opens in Photoshop as a Smart Object. 2.
I n the example of the d rive-through wedding chapel photo, the goal of the fi rst raw conversion is to optimize the overall appearance of the image without cli pping any of the highl ights. Click the word Default to use the Default setti ngs as a s tarting poi nt. A d j ust the Recovery slider to +20 to reduce the small amount of h ighl ight clipping along the far edge of the awning (Figure 6-96) and click the Open Object button to bring the file i n to Photoshop.
Figure 6-94. The original rawfile (top) was processed twice using different settings to emphasize different parts of the image. The two versions were combined using Smart Objects and layer mas/(s to create thefinished version on the bottom. s/)
'1J c h06_WeddingCh apel.dng 1.
Open the file i n Adobe Camera Raw. Before we make any adj ustments to the i m age, click on the blue workflow s ummary link at the bottom of the dialog to open the workflow options and click the check box for "Open in Photos hop as Smart Objects". Set the other settings as s hown in the i l l ustration and click OK to close the Workflow Options dialog (F igu re 6-95) .
Figure 6-96. Thefirst processing ofthe rawfile is designed to control the bright highlights. 3.
Once the file is open in Photoshop, choose Layer-> Smart Objects ->New Smart Object via Copy (you can also right-click the layer and find the same choice in a contextual menu). This will c reate a copy of the fi rst Smart Object that can be adj us ted independently.
Placing the raw file as a Smart Object in the layered file allows you to double-click the layer t h u mbnail at any time to reopen the Camera Raw d i alog and adjust the setti ngs. It's essenti a l l y l i ke h aving a raw fil e a d j ustment layer.
Contrast and Exposure Control I
209
4.
Double-click the thumbnail for the d u plicate Smart Object to open the Camera Raw dialog. Adj ust the file so that d a rker shadow areas are lightened. H ighlight regions may start c lipping but don't worry about that. In our adj ustments to this i m age, we set the exposure to + lAO and the brightness to + 1 00 to reveal more of the angelic cherub details i n the awning. We increased the vibrancy to + 2 5 to punch up the colors a bit and set the Clarity to 50 to enhance the midtone contrast. Click OK to apply the changes to the Smart Object (Figure 6-97).
In Camera Raw, the Option/Alt key will change the Open Image/Objects button to Open Copy. This will open a copy of the file with the current settings, but it will not update the metadata that contains the Camera Raw processing instructions. This is useful if the settings you are applying are temporary as in cases where you are processing multiple versions of the same raw file with different interpretations. Pressing the shift key temporarily changes the Open Image button to Open Object, allowing you to create Smart Objects as needed rather than changing the default behavior.
Figu1'e 6-98. Two difef rent intelpl'etations of the same raw image stacl(ed as SInCl1·t Object layers in a singlefile.
The first part of this process, opening a raw file as a Smart Object, then making a copy of the Smart Object and processing the raw file to emphasize different tonal areas in the image, is basically the same for all photographs. The exact nature of your raw adj ustments will vary, of course, as will the placement of the Smart Objects i n terms of whether the lighter or darker version is on top, but the basic procedure is the same. It is the type of layer masks that you use, and the selection techniques used to create and edit those masks, that will be different for each photograph. The layer mask for the d rive-through wedding chapel photo is a pretty simple one. 6.
Choose the Quick Selection tool (W) and tap on the right bracket key until you have a very large brush; for the 3072x2048 image in this example, we used an 800pixel brush. Drag from left to right across the top part of the image to create a selection of the top half that contains the blue awning (Figure 6-99).
7.
Now tap the left bracket key to make the brush size m uch smaller so you can use it in the blue area above the d rive-through window that was not selected in the fi rst pass of the Quick Selection tool. We used a 3D-pixel brush. Drag over this area to add it to the selection (Figu re 6- 1 00). There m ay be some a reas of the selection that need touching up but that can be handled later.
Figure 6-97. In the raw processing ofthe Smart Object copy, the goal was to reveal the darker shadow detail in the top ofthe awn.ing. 5.
210
Double-click i n the Smart Object layer names and rename the lighter version " Overall Exposure," and the darke r version "Highlights."
I Chapter 6 : Dodging, Burning, and Exposure Control
Figure 6-99. Using tl1e Quicl( Selection tool and a lcllge brwh, drag across the top part ofthe image to select it. 9.
To soften the edges of the mask, choose Filter->Blur-> Gaussian Blur. For this photo, we used a bl ur radius of about 2.5 pixels (Figure 6- 1 02) . The amount of blur you use wi ll vary depending on the size of the image, the exact nature of the mask you are working with, and how precise the mask edges need to be. We also used the Brush tool and painted on the layer mask with white at 50% opacity to reveal more of the lighter version in the area of the kissing cherubs and the d rive-in wi ndow sign (Figure 6- 1 03).
Figure 6-1 00. Adding to the selection with the Quid( Selec tion tool.
8.
Make sure that the top Smart Object layer (the lighter version) is active in the Layers palette, and click the Add Layer Mask icon at the bottom of the Layers palette to create a layer mask that shows the selected area but hides the lower part (Figure 6-101).
Figure 6-/01. The addition oftl1e layer mask creates tl1e initial blend ofthe two Smart Object layers.
Contrast and Exposure Control I
211
10.
As a fi nal step, we selected the "Please Pull Forward" sign with the Polygonal Lasso and added a Curves adj ustment layer to increase the contrast of the sign. We also painted on the first layer mask with white at 25% opacity to lighten the d riveway a bit (Figure 6- 1 04).
� � �Prtvitw
Figure 6-102. Applying a slight' Gaussian Blur to the layer masl( di;guises some ofthe hcml edges.
Figure 6-104. The Curves acy'ustmentfor the "Pull Forwcnd" sign and thefinal cO�11posite ofthe two intelpretatiollS ofthe same rawfile. .\1, '
The advantage of placing a raw file as a Smart Object is that you have the ability to revisit the raw conversion at any time without redoing any of the additional layers that you may have added after the original raw conversion. The raw data is embedded in the layered Photos hop file and is not l inked i n any way to the location or name of the original raw file. The only thing to keep i n m i n d is that any local tonal corrections might need to be updated if you decide to go back and revi sit the settings of the initial raw conversion. Combining multiple exposures with l ayer m asks
Flgure 6-1 03. The layer mask was modified wing the brush tool to show more of the lighter version over the cherubs and the drive-in window sign.
21 2
I Chapter
6:
Dodging, Burning, and Exposure Control
When you're facing a scene with high dynamic range and you have a tripod with you, photograph one exposu re for the highlights, midtones, and shadows without changing the aperture. I n Photoshop combine the d i fferent i mages as layers and blend them together with layer masks. Tn this technique all of the exposure modification is done in the camera with the capture of different exposures designed to opti mize a specific part of the tonal range. After opening the images, the basic layer masking steps to composite the images together a re the same as those used i n the example of the wedding chapel photo. The masks you use, of course, as well as the methods used to create them, will be specific to each image.
Overall Exposure
Exposure for Shadows
��-----------------------,
Combined Exposure
Figure 6-105. Two tripod-mounted exposureJ, one for the overall scene and one to show the shadow details, combined into a single image using layers and layer masks. u Sf)
Open both files into Photoshop and arrange them side by-side. Make the lighter version the active file. Select the Move tool, click in the lighter image, hold the Shift key down and d rag it over onto the darker version. Keep the Shift down until a fter you have released the mouse button. You should now have the lighter version of the i mage as a new layer on top of the darker version. Rename the layers: "Shadow Detail" for the lighter image, and "Overall Exposu re" for the darker version (Figure 6- 1 06).
Holding the Shift key down as you drag one file onto another will ensure that the image you are dragging is centered in the target file. When both images are the same pixel dimension and they are two exposures of the same scene, as in this example, they will be perfectly aligned.
Figure 6-106. The two exposures ofthe lighthouse arranged as I({yen in a single file.
Contrast and Exposure Control I
213
2.
With the S h adow Detail layer active, choose Layer---> Layer Mask--->Hide A l l to add a black layer mask. The ligh t version o f t h e i m age will be hidden, revealing the darker i m age below it (Figure 6- 1 07).
3.
Choose the Brush tool and select a soft-edged 200-pixel brush tip from the B rush Picker in the Options Bar. Press "D" on the keyboard to set the default colors, which wil l make w hite the foreground color. In the Options Bar, set the brush Opacity to 30%.
4.
Make sure the layer mask is active (look for the highlight border a round it) and brush over the areas of the i m age where you want to show the lighter shadow detail from the top layer (Figure 6- 1 08) .
5.
If you have revealed all of the brighter rocks and you still would like it to be a bit l ighter, set the blending mode for the top layer to Screen. This will cause the area of that layer that are visible in the composite image to become even lighter (Figure 6- 1 09). Fine tune the effect with the layer opacity slider i f needed.
Figure 6-1 08. The modified layer mask reveals the lighter shadows tones ofthe top laye;:
Figure 6-1 07. A black layer masl( hides all ofthe lighter exposure.
Figure 6-1 09. Setting the blending mode for the top layer to Screen will lighten the mcks even more.
214
I Chapter
6:
Dodging, Burning, and Exposure Control
Combining Multiple Scans of a Single Negative I f you are working with a high-contrast film image and are having d i fficulty capturing the ful l range of tones i n single scan, y o u can u s e a v ariation of the m u l tiple process raw conversion technique. In order for the different scans to al ign properly in Photoshop, the slide or negative, as well as the crop box you define i n the scanner software, cannot be moved once you start the process. Make two scans and use the scanner software to change the settings for each scan so that d i fferent areas of the tonal range are optimized (for example, a good overall scan with no highlight clipping and another one that lightens
Combining multiple bracketed exposures with Merge to H D R H D R stands for high dynamic range, a n d is used to blend a series of camera raw exposures i nto a single i mage that uses the best values of the different source exposures to create a final image with a d ynamic range that would be i m possible to capture with a single shot. In scenes w here there is an extreme contrast range with dark areas that need to be visible and very bright highlights, this method can create impressive results. Photographic considerations for HDR photography
The key to a successfu l H D R blend lies in how the photographs are made. A tripod is mandatory to minimize camera movement and an electronic cable release is also a good idea to avoid inadvertent v ibration when the shutter button is pressed. Finally, i f you are using a lens with a vibration reduction feature, it's best to turn this off for the tripod-mounted exposures. And if your camera has a mirror lock-up control, that's yet another way to reduce noticeable vibration, especially at slower shutter speeds. In terms of camera settings, your exposures should ideally be one flstop apart, and the various flstops should be achieved by changing the shutter speed and never the aperture. To keep the focus and size relationships consistent, the aperture needs to be the same for all the exposures. The number of shots needed will vary depending on how extreme the contrast range is in the scene you are photographing, but three to seven shots is a typical number of exposures (Figure 6- 1 1 0).
the shadow values). For the best quality, use the scanner's h igh bit-depth capabilities to scan at more than 8 bits per channel. Combine the two scans into a si ngle fi l e using the steps shown in this chapter (see "Multi ple raw p rocessing of a single exposure"). If you have m ultiple slides or negatives that were shot o n a tripod with bracketed exposures, the success of aligning separate scans that are sized identically can be accomplished with the new CS3 Auto A lign feature. File-Scripts-Load Files into Stack and check " A ttempt to Automatically A lign Source I mages."
The one thing that HDR photography does not do well is motion . While using a tripod can keep the camera stable, motion i n the scene is beyond your control. If there is significant movement i n the scene, such as trees swaying in the wind or moving clouds, then this could cause blending irregularities and ghosting in the fin ished result (Figure 6- 1 1 1). In a classic landscape photograph, this is probably not a n acceptable side e ffect for most people. For scenic elements such as rivers or oceans you can sometimes cover up the blurry, ghosted section by using a single image from the series with a layer mask after the i nitial H D R merge has been completed. As tricky as motion can be, however, it's im portant to note that for some types of photograp h y such blending "mistakes" a re not necessarily a bad thing. I n some cases, they can look very interesting, especially with people or bright lights moving through the shot and leaving strange ghosting effects as they pass. In the H D R photo of taxis leaving an underground garage (Figure 6- 1 1 2), the transparent trails and saturated color artifacts created by the moving cars convey a sense of the hectic activity at a busy Las Vegas hotel.
Contrast and Exposure Control I
215
Figure 6-1 ZO. To capture the full range of tonesfrom the deep interior shadows to the detail in the sunlight-washed seat of the chail; six exposul·es were made. The aperture for all six shots wasfl8 and the shutter speeds rangefrom 1I400th of a second down to 1113th of a second (ISO 200). Merge to HDR pre-fl ight check
Before you start the Merge to HDR process, there are a few things to double-check. Fi rst, i n Photos hop the HDR results will be best when you use raw files from d igital cameras. This is because the raw files a re a l ready captured at a h igh bit depth, whereas JPEGs a re l i m i ted to 8-bits of data. It does not work with scan ned i mages. Second , the raw files should have had no tonal adj ustments applied to them i n Camera Raw. Fortunately, there is a "secret handshake" between Adobe Bridge, Adobe Camera Raw, and the Merge to HDR command that negates any previously applied adj ustments when accessing the HDR command th rough Bridge.
Figure 6-111. The clouds were moving during the exposure of the source images for this HDRfile, resulting in obvious alignment problems in thefinalfile.
Before you sta rt with Merge to HDR, you should also double check the image size in the Workf1ow Options of Camera Raw. This setting is sticky, meaning that it will al ways use the size that was specified the last time you processed a raw file. Double-click a raw file in Adobe B ridge to open it i n Camera Raw, a n d check the Workf1ow S u m ma ry i n the hyperlink at the bottom of the dialog. If the size is not what you want, click the link to open the Workf1ow Options, choose the correct size, click OK, and then click Done in the main ACR dialog. Since Merge to HDR can be a memory intensive process with large files, you may want to consider using smaller versions d u ring the i nitial proofi ng process to see how well the merge works. If all looks good, then you can redo the merge using the full-size files.
� ch06_BodieChair (fi l es
1 through 6)
Accessing Merge to HDR
Figul·e 6- 112. For some scenes, ghosting and altgmnent irregularities caused by motion in the J"("ene l�· not a problem, and can even enhance the image. This is a blend of six exposure,-. , \/J 216
I Chapter 6: Dodging, Burning, and Exposure Control
You can i ni tiate the Merge to H D R process from Photoshop or from Adobe B ridge. From P hotos hop, choose Fil e--> A utomate-->Merge to HDR. In the i n itial dia log, specify which files you want to use and click OK. We think it's easier to do this from Bridge, however, because you can easily see the thumbnails of the i mages you want to use. Simply select them all by clicking the fi rst one and then Shift-clicking the last one. Choose Tools-->Photoshop-->Me rge to HDR. Photoshop will then open all of the fi les and load them into the initial Merge to HDR dialog. Depending on how
many files are being loaded, their size, and the speed of your computer, this can take several m i nutes. Step 1: Merging the source files. Merge
HD R is really a two-step process (actually, it's more of a three-step process since you al ways have to apply further fine-tuning after the HD R conversion). The first step involves creating a 16 - bi t or 3 2 - bi t fi l e that is a blend of a l l the source files you have chosen. The initial Merge to HD R d ialog displays thumbnails of the source images and their exposure value (EV) on the left (Figure 61 1 3). You can remove an image by clicking in the checkbox. I deal l y YOLl want to LIse only enough files to cover the visible dynamic range in the scene. The more files YOLl use, the greater the l ikel ihood of m i s a l ignmen t and de ta i l softening.
Merge to HDR: 32-Bit or 16-Bit?
to
"$-=.---•. o
I ( --�- .. -)
On the right side of the Merge to H D R d ialog box is a menu for setting the bit depth of the merged file. The default setti ng i s 32 Bit/Channel. For photographic i mages that w i l l be pri n ted or displayed on screen, a 32-bit fil e is more than you'll ever need. The ability to generate 32-bit files is i ntended for 3D i maging programs, such as those used i n movie special effects or the creation of computer game graphics. If you choose to create a 1 6-bit file, then you will be taken d i rectly to the H D R Conversion dialog box, w here you can apply further modifications. For photographers, the primary reason to process the file as 3 2 -bit is that you can save the 32-bit version of the file, which can then be processed in d i fferent ways for the final HDR conversion. Si nce the most time-consuming part of the H D R process is the i nitial merging that takes you to the Merge to HDR dialog, we think it's a good idea to be able to save the file for further H D R i nterpretations without starting t h e entire process over again . For this reason, we choose to create a 32bit file and then save that in the Portable Bit Map format (pbm ) or Photoshop (psd) formats.
Figure 6-713. The Melge to HDR dialog box. On the rig/It is the bit depth menu and the white point preview histogram.
The initial merged file that is previewed contains a tonal range that is too wide to be d isplayed on your monitor at one glance. You can use the slider under the h istogram to preview the d i fferent sections of the tonal range as well as set the white point for the final merged i mage. Each of the red ma rks under the histogram represents one stop or exposure value (EV). If you move the slider to the left you can see how much shadow detail the file contains (si nce the dynamic range is wider than your display capabil ities, the highl ights will blow out when prev iewing the shadow details). We usually prefer to set the white poin t preview so all of the highlight detail is visible with no c l ipping. With this setting, the previewed image w i l l look quite dark si nce the monitor cannot display the full range, but the detail will be preserved in the highlights and detail will sti l l be retained in the darker areas, even though you cannot see it (Figure 6 - 1 1 4). Balancing out the tonal values and revea l i ng the shadow detail will take place i n the second part of the process. Click OK to create the merged file.
..- �
L' • - ...1 I
--
-=.- e0·--'
.
, ..
_ ,
Figure 6-114. The white poillt preview slider i.i moved all tile way to the right side of the histogram to show all of the bright highlight detail.
Contrast and Exposure Control I
217
� � �Prevlew
Moving the White Point Preview slider affects only the preview of the image. As long as you use the 32-bit option, all of the HDR data is preserved in the merged version of the file. When you open the 32-bit file in Photoshop, the preview adjustment is stored in the file but can be changed at any time by choosing View-..32-Bit Preview Options. This dialog has an Exposure and Gamma slider that can be used to open up more shadow detail prior to the final HDR Conversion step.
When the 32-bit file appears, it will look very dark, like the preview in the Merge to HDR dialog box (Figure 6- 1 1 5). But fear not, more options to tweak the appearance of the image a re j ust a step away. The fi rst thing you should do is save the file so it can serve as a source for further explorations if you want to try i nterpreting it in different ways in the HDR Conversion dialog box. B y saving the file you don't have to go through the time-consuming fi rst part where the source files are merged into a 32-bit file. Choose File-.. Save As and, in the Save dialog box, open the format menu and choose Portable Bit Map or Photoshop. A fter saving the file, you can also choose View-..32-Bit P review Options to fu rther adj ust the preview of the merged file (Figure 6-1 1 6). Remember that the tonal range of a 32-bit file is much too wide to d isplay on a monitor, so all you are doing here is adj usting the preview of the file. The sliders i n this dialog box provide considerably more control than the Merge to H D R dialog for getting the preview looking good .
Figure 6-1 1 6. Using the 32-Bit Preview Options gives you more flexibility in adjusting the preview of the 32 -bit n2eJgedfile. Step 2: The HDR Conversion dialog box. The cool part of the Merge to HD R process comes in the next step, when the image is converted to a 16-bit-per-channel file. Unfortunately, when you start from a 32-bit file, there is no hint that further action is requi red. I t would be nice if a message dialog appeared that informed the user that additional options could be found when the file is converted to 1 6-bit.
To allow for further conversion flexibility and the opportunity to return to the original 32-bit PBM file, choose I m age-..Duplicate, and then choose I mage-..Mode-.. 1 6 Bits! Channel to bring up the HDR Conversion dialog box. O f the four conversion methods available, only Exposure and Gamma and Local Adaptation give you any sort o f control over how the image looks. The latter method is the one we use most of the time but we'll take a quick look at what the others offer. Exposure and Gamma
Figure 6-1 15. This melged 32-bitfile is very dark because the white point preview wa,' set to show all the detail in the brightest highlights. 218
I Chapter 6 : Dodging, Burning, and Exposure Control
These sliders are the same as 111 the 32-Bit P review Options and i f you adj usted the image there before converting to 16-bit, then the sliders will show the previous settings. The Gamma slider will increase or reduce the contrast i n the image and functions similar to an overall midtone brightness control. The Exposure slider controls the overall brightness of the image in a way that is much more abrupt; you don't have to move it very much to see a d rastic e ffect on the file. These controls are fairly blunt and i n most cases we don't find them too useful (Figure 6 - 1 1 7).
:1 !=WI [i7sI
� � � �
:1 There are no oplions for thl� (onverslonmelhod
� PJeview
Figure 6-J J 7. The Exposure and Gamma sLiden allow for basic brightness and contrast adjustments but not much fine>'se in how those changes are applied to the image.
!S!I Prevlew
Figure 6-J 18. /-hghlight Compression is an automatic adjustment with no way to modify the effect. :1
Highlight Compression
This is an automatic setting and there are no options for mod i fying the e ffect (Figure 6- 1 1 8). I t compresses the h ighlights so they fal l within the brightness range of the 8-bit or 1 6-bit file. While this method can open up the mid tones a bit while still preservi ng all of the bright highlight detail , the shadow tones usually remain very dark.
� � � �
Thue are no opllon!o (or lhll convenlon melhod
� � S Pre�
Equalize Histogram
This control will remap the brightest h ighlight and da rkest shadow points to the normal comrast range of a low dynamic range image (that is, 8- or 1 6-bits). The result here is usually better than either Exposure and Gamma or Highlight Compression, and i t can sometimes look i nteresting, but as it offers no options to modify the effect, this method also leaves a lot to be desi red (Figure 6- 1 1 9). Local Adaptation
This method gives you the most con trol and is the one that we use for nearly all of our HDR conversions. Click the triangle button to expand the dialog and reveal the tone curve and histogram. This is where the real conversion fine-tu n i ng takes place.
Figure 6-1 1 9. Of the two automatic adjustments in the HDR Conversioll dialog, Equalize Histogram is the most interesting, but it's >·till a formula-settingfeature with no fine-tuning available.
Contrast and Exposure Control I
219
Before you ad j ust tbe tone curve, the fi rst step is to set the Radius and T h reshold sliders. Radius controls the size of the local brightness areas, and Th reshold specifies how much difference there must be between neighboring pixels' tonal values before they are no longer included in the same brightness area. For many i mages, the default Radi us value of 16 pixels is too h igh. Start with a fairly low threshold of 0.5 or 0.75, and then adjust the radius until the blend i ng along the contrast edges looks more natural. As with the Shadows/Highl ights command, which contains similar tonal separation functionality, there is no right setting to use. The important thing is to understand how the sl iders a ffect the image and w hat artifacts to be on the lookout for. The main thing to be wary of with these controls is the creation of noticeable halos a round high-contrast edges (Figure 6- 1 20 and Figure 6- 1 2 1 ). Zoom in to 100% (View->Actual Pixels) for the best view of the edge details.
:\
:\ � --00----
[o.75l
-�Gl__---
� � � �Prtvie.w
� � � � �Preview
Figure 6-121. A lower Radius value results in better-lool(ing contrast edge>'. Adjusting the Local Adaptation curve. The h is togr a m
represents the brightness values of the origi n a l HDR image, and in many cases it will not span the entire tonal range (the exam pl e of the old chair does not have many values on the dark shadows). The red marks along the bottom of the cu rves grid represent increments of 1 EV (approximately 1 f-stop).
Figure 6-120. The default Radius setting produces lund-edged halos on the bright contrcw edges in this image.
220
I Chapter 6 : Dodging, Burning, and Exposure Control
The object in this dialog box is not to end up with a perfect image in terms of brightness, contrast, and color saturation, but to create a version that has the ful l range of useable tones that you can work with in Photoshop once the 32-bit file has been converted to 1 6-bits. If there a re no tonal values on either end of the histogram, then i t's usually a good move to bring the highlight or shadow points in to j ust before where the tonal data starts. To ensure that the darkest shadow tones still hold visible detail, we placed two points on the lower section of the curve and raised them u p a bit ( Figure 6- 1 22).
� � � � !ll "_
Figure 6-122. Fine-tuning the Local Adaptation curve.
The default behavior of the curve is to try to equalize you r changes across a l l the points. I f you click a point to make it acti ve, and then click in the Corner checkbox, the point becomes a corner poin t and no equal ization will be performed when you insert and mani pulate another point (Figure 6- 1 23). The simplest way to think of the corner poi nt is that it provides a n easy way to lock down a section of the cu rve so that it does not move. But i f the changes to the curve become too extreme on either side of the corner point, then you run the risk of introducing color posterization or hard halos along high-contrast edges.
Figure 6-123. You can use the Cornerfeature to tramform a normal control point into a comer point.
Of course, i f you're after a non-traditional look, don't be shy about doing wild things to the curve. You can create some pretty interesting color solarization effects simply by inverting points on the curve (that is, raising shadow values up and lowering highlight poi nts; see Figure 6- 1 24). I t may not be what Ansel Adams vvould do, but Man Ray would certainly approve !
Click the Save button t o save the tone curve setting to apply to other images or to reinterpretations of the saved 32-bit pbm version of the file. Use the Load button to access your saved HDR settings.
Figure 6-124. A Melge to HDR homage to Man Ray.
Contrast and Exposure Control I
221
Step 3: Further adjustments to the 1 6-bit merged file.
Once you click OK in the HDR Conversion d ialog, your adj ustments are applied to the 32-bit version of the file, and it is converted to 1 6 bits (or 8 bits i f you chose that option, which is not recommended). The results from the HDR conversion are typically f1at and low contrast (Figure 6- 1 2 5). If you regard this as the finished product you're l i kely to be d isappointed. At this point you will almost always need to apply further modifications using adjustment layers and layer masks to i mprove the contrast and sometimes the color balance. In the image of the old chair, we added a Curves layer to slightly increase the contrast and warm the image up a bit, a Selective Color layer to enhance the density of only the black tones, and a final Hue/Saturation layer to slightly desaturate the image (Figure 6- 1 26).
Figure 6-126. Thefinal version after adding some additional adjustment layersfor contrast and color correction. The wide dynamic range present in the original scene could never have been recorded with a single exposure. () Sf)
Figure 6-125. The initial resultfi'om the HDR Conversion dialog has a much fuller range of tones than any ofthe original source exposures, but it�· velYflat and dull.
222
I Chapter 6: Dodging, Burning, and Exposure Control
Combining m u ltiple HDR images
I n some cases, the results from a single Merge to HDR may not be exactly what you want, and you may choose to combine two d i fferent HDR interpretations of the same file. Once you are fi nished with the HDR part, this is essentially the same technique as combining two i nterpretations of the same raw file. The type of masks that you use will be determined by the needs of the images you are working with.
He combined the two files i nto a singl e file as separate layers and used layer masks to blend the d a rker sky version with the one that revealed detail on the steep mountains. He used several additional adj ustment l ayers with layer masks to further e n hance each of the two separate images until he created a final version that better expressed h i s memories of watching the sunrise o v e r t h a t magnificent place (Figure 6- 1 2 8).
Sean fou n d the multiple Merge to HDR method quite useful in creating an image of the sun rise over Machu Picch u in Peru. He started with six exposures all recorded at one f-stop apart. The aperture was a constant fl8 and the shutter speeds ranged from 1 I3200th down to 1 I 1 00th of a second. After creating the i nitial 32-bit H D R fi le, he saved it in the Portable Bit Map format so that he could try out d i fferent i nterpretations without having to redo the time consuming merge process. A fter finding that a single file was not yielding the result h e envisioned, he created one 1 6bit conversion that emphasized the detail in the mountains and another that was optimized for the sky and rising sun (Figure 6- 1 27).
Figure 6-127. The six source exposures of dawn at Maclw Picchu and the two 16-bit HDR conversions tilat were combined using layers alld layer masks.
Contrast and Exposure Control I
223
Figure 6-128. Madnt Picchu Dawn: a blend ofsix exposures, two HDR conversions, and numerous adjustment layers
.
S e e i n g th e Light In this chapter, w e have covered many techniques for manipulating the tonal values i n a photograph. Although knowing the software specifics ofthese techniques is necessary for applying them to an image, perhaps even more important is a skill that is totally separate from Photoshop or even the computer. That skill is recognizing the crucial role that brightness val ues-the delicate interplay between light and shadow-play in a photograph and how the mani pulation of those tonal values can d ramatically transform an image. Learning how to see the tonal possibilities that are present in a photograph and k nowing how to control these qualities through creative dodging, burning, and exposure blending is essential to creating the i m age you want.
224
I Chapter 6 : Dodging, Burning, and Exposure Control
) . Sf)
Figure 6-129. Temple ofthe Sun, Machu Picchu. This is a 6.3 minute raw exposure taken on a very dark moonless night. A smallflashlight was used to illuminate the ruins by ''painting'' them with light. The reddish glow in the skY is from the lights ofa dam a few miles dowll rive!: The image has been enhanced with minor dodging and buming. Levels), Before you proceed any further, double-check that the sample a rea for the eyedropper is set to a n appropriate size for working with photographs. Move your cu rsor over the image and right-click to open the eyed roppe r contextual menu. Choose 3 by 3 Average, which is a good size to use for photographic work (Figure 7-24).
)
D� current
eK==:::: 1: (�===
(
Cancel
( Add To Swatches ) ( Color libraries O L: � O a: C]
O b: C] c: D "
M: D " y: D " K: IU" Figure 7-25. The tCliget valuesfor a neutral highlight tone.
Global Color Correction I
237
Select target shadow color:
(
When yo u release the O pt i o n/Alt key, t h e c l i p p i n g d i splay goes away. To h e l p pi n po i nt the t o n a l area
OK
( Cancel ( Add To Swatches )
so you can p l ace a color sam p l e r th ere, or move the Leve l s d i alog box, and use a corner of it as a
Color L i braries
" poi nter" to show yo u w h e re the b r i ghtest tonal
O 11 I O a: Ll O b: 10 I C: �% M: � % Y: E1 % K: � % L
point i s ( F i g u re 7-28). O n ce yo u have a c o r n e r of the d i a l og box pointing to the tonal area yo u want to m a rk, it's m u ch eas i e r to locate it aga i n .
Figure 7-26. The talget values/or a neutral shadow tone.
Near the e n d of t h i s exe rcise a message d i a l og box w i l l appear after you c l i c k OK i n Levels asking if you want to save the n ew target co l o r as defau lts. G o ahead and c l i c k OK to this si nce these val ues are good ones to use for photographic work. It's sti l l a good idea to d o u b le-check the eyed ropper target values before u s i n g them, especially i f you work in an e n v i ro n m e n t where m u ltiple people use the same computer.
I d e nti fy i n g the brightest and d a rkest poi nts 6.
238
To identify the brightest point, hold down the Option! Alt key, and click the I n pu t Levels h ighlight slider to activate the clipping d isplay. The i mage will turn black. Keep the Option!Alt key down and slowly move the slider to the left until parts of the image begin to appear in bright colors. This indicates the brightest a rea i n either a single chan nel or i n two channels. The point where the i mage turns white fi rst reveals the tone in the i mage that is the brightest in all three color channels. In this photo, it's no surprise that the white clouds are the brightest point i n the image (Figure 7-27).
7.
Release the Option/Alt key and press Cmd/Ctrl + (plus) several times to zoom in for a close view of the area i n question (press the spacebar and reposition the image if necessary). Press Option/ Alt and the highlight slider agail) to reveal the brightest point.
8.
Shift-click the brightest point for the highlights to place a color sampler point there. This will appear in the I n fo palette as #1.
I Chapter 7 : Color Correction
Figure 7-27. The highlight clipping display i n Levels showing the brightest point in the Mt. Silliman image. 9.
I f you need to reposition a sampler point a fter you have placed i t, move the cursor d irectly over it, and, when the a rrow appears, S h i ft-click i t to d rag it to a different position. To remove a sampler poi n t while you are stil l i n the Levels d i alog box, S h i ft-Option/Alt click i t.
10.
Retu rn the highlight slider back to the fa r right side of the Levels h istogram. Now zoom out (Cmd/Ctrl minus) and repeat the procedu re described in steps 6 and 7 to identify the darkest tone in the i mage. The only difference is that you will be accessing the shadow cli pping display by Option/Al t-c1icking the I n pu t Levels shadow slider. Once you have found the darkest tonal value in all th ree c hannels (in this image, it is in the trees i n the lower-left side of the photo), Shift-click it to place a second color sampler poi nt (Figure 7-29).
� ( ( (
( ( ( (
) ) ) ) ) � ( ) I/I"I�I
6K Reset load . . .
1m I
Channel:
(
RGB
Input Levels:
•
•
I
Output levels:
10
I
11.00 I
6
1m I
11.
load . . . Save . . .
•
Auto
Options . . .
10
•
6t( Cancel load . . . Save ... Auto Options ...
} ) ) ) ) )
I
� Preview
Figure 7-29. Top: Finding the darkest point in the shadows ofthe forest in the lower-left side ofthe image. Bottom: Placing a color sampler point at that location. 13.
Click the h ighlight Eyedropper tool to make it active. Move the precise crosshair c ursor over the existing sample point # 1 . When they line u p perfectly, click once to remap the tona l v al ues of that point to the new target val ues of 248, 248, 248. If you accidentally click i n the wrong place, simply realign the c ursor with the sample point and click again . When you have successfu l l y set the highlight point, click the highlight eyedropper agai n to turn i t off.
14.
I f you check the sample points in the I n fo palette, you should see two sets of numbers separated by a slash. The n umbers on the left are the original values and those on the right are the adjusted values (Figu re 7-30).
Neutralizing the highlight and shadow points 12.
( ( ( ( ( 6 � (
Output Leve ls:
•
51 Preview
I m portant: After placing the color sampler points, remember to return the Levels sliders back to their starting positions. These adjustments were made only to reveal the brightest and d arkest points and are not somethi ng that you will apply to the i mage.
! : I----�
RGB
I�I
Cancel
Figure 7-28. Top: Using a corner of the Levels dialog box to "point" to the brightest spot. Bottom: a color sampler point is placed to tracl( the values of that precise area in the Info palette.
(
Input levels:
{ ( ( ( ( 0 I ( I� I?'I�I"'I
i �I
•
� Preview
I
Channel:
10
Save ...
Options ...
6
•
load . . .
Auto
Save ...
Output level s :
10
6K
Reset
Novv zoom back in to 1 00% (View-Actual Pixels) to see the h ighlight point that you marked with color sampler poin t #1 . Press the Caps Lock key to set the cursor to precise mode (a crosshai r).
Global Color Correction I
239
_ x
ItfTnfOXL
�
R:
2091 251
c:
191
1%
J! G : B:
2231 252
J( M : Y: K:
41
0%
2 2 61 2 5 0
+.
81
1%
01
0%
8-bit
8-bit
X: Y:
1080 639
I W r.:r
H:
#l R :
2071 248
#2 R :
331
40
J( G : B:
2 2 11 2 4 8
J( G : B:
3 41
38
301
34
2 2 41 2 48
Doc: 25.0M/25.0M
FigU1'e 7-30. After clicl(ing highlight color sampler with the white point eyedroppel: The before and after valuesfor the clouds highlight point are visible in the Info palette. 15.
Now zoom i n on the area where you placed color sampler poi n t #2 for the darkest poin t i n the image. Once again, make sure that the zoom view is 1 00% (View-->Actual Pixels).
16.
Click the shadow eyedropper to m ake it active, al ign the cursor over the sample point as in step 1 3 , and, once they a re aligned, click once to remap the shadow poin t's tonal val ues to the new target values of 1 0, 1 0, 1 0. Once this is accomplished, click on the shadow eyedropper to turn it off.
17.
Check the I n fo palette to see the adj usted shadow val ues for sam pie poin t #2 (Figure 7-3 1 ). x
K:
c: J! M ' Y: K:
8-bit
8-bit
+.
Remapping the highlight a n d shadow poi nts to k nown neutral target values with the eyedroppers does two things. First, it sets the brightness value of the highlight and shadow point, which can often help im prove the brightness and contrast of the image. Second, i f a color cast is a ffecting the high lights or the shadows, it can be removed using this procedure. I n the case of the Mt. Silliman photo, brightness, contrast, and overall color balance were greatly improved by this process. I f a color cast is still noticeable after usi ng the eyedroppers, then it's l ikely that it's l u rking in the midtones and will have to be add ressed with a n additional correction.
.;
(IiifOXl. J(
Figure 7-32. The Layers palette showing the overall Levels adjustment and a Curves layer tlnlt slightly lightens the top ofthe mountain.
Good and Bad Color Casts
I W
x:
r.:r
Y:
H:
#1 R :
2071 247
#2 R:
331
10
J( G : B:
2 2 11 2 4 7
J( G : B:
341
10
301
10
2 2 41 2 4 7
Doc: 25.0M/25.0M
Figure 7-31. After clicl(il1g on the distantforest with the black point eyedroppel: The before and after valuesfor tileforest shadow point can be seen in the Info palette.
The shadow areas at the top of the mountain became a bit too dark from setting the shadow poi nt, so we added a cu rves adj ustment layer with a layer mask to lighten only that part of the photo (Figure 7-3 2). This adj ustment to the top of the mountain was merely some additional fi nessing. Most of the significant i mprovement i n the Mt. Silliman image was created by the simple step of neutralizing the h ighlights and the shadows.
240
I Chapter 7 : Color Correction
Using this method to neutral ize the h ighlight and shadow points in a photograph can also be achieved using the Curves d ialog box. In fact, with the changes to C urves that were introduced in Photoshop CS3, portions of this procedu re a re easier i n Curves than in Levels.
Before you venture too far i n to the neutral zone, one thing to keep i n mind is that not all color casts a re bad. I ndeed, for some photos, a specific color bias might be a key element in the i m age (for example, the yellow glow of fi relight, performance shots illumi nated by colored stage lighting, a s unset photo i n fused with the orange pink glow of day's end). There a re also times when an i m age just doesn't look right a fter you have neutralized the key tonal a reas, and you may need to adjust it so that the color balance falls somewhere between a strict neutral and the original cast that you tried to remove. Neutralizing key val ues such as the highlights, shadows, and, in some cases, specific mid tone v alues, is a very useful technique for removing color casts, but i t is not always right for every Image.
Setting neutral h igh lights and shadows with Curves The Cu rves dialog box in Photoshop has long had the same Eyedropper tools as the Levels dialog box, but until CS3, using them to set neutral highlights and shadows has not been as efficient because the dialog box had no clipping display. I n CS3, the C u rves dialog box received a significant upgrade, and one i tem of new fu nctional i ty is the abil ity to turn on the same clipping d isplay as i n the Levels d ialog box. The Cu rves cl ipping display even offers some key fu nctionality that Levels does not: The cl ipping d isplay remains visible without having to hold down any keys or click any sliders or control poi nts. This makes it m uch more efficient to evaluate the clipping and place the color sampler points than in Levels. Here's a n overview of how to use the eyedropper neutralizing technique in Cu rves (Figure 7-33).
the cu rve as in Levels to show highlight and shadow cli pping). 2.
Shift-click on the brightest a rea (the place that turns white fi rst) to place a color sampler point there. A fter the sam pler point is placed, return the highlight control point to the upper right corner.
3.
To show the shadow clipping, click the shadow poi nt (lower-left corner) and move i t horizontally along the bottom of the grid to the right. Keep moving the point to the right until you begin to see cli pped a reas. The place that turns black first is the da rkest point i n the image i n all t h ree channels. Shift-click to place a color sampler poi n t on the da rkest a rea (Figure 7-35) .
4.
Select the white point eyedropper and click on the highlight color sampler poi nt to neutralize the highligh ts. Select the black poi nt eyedropper and do the same for the shadows color sampler poi nt to neutralize the shadows (Figure 7-36).
Figure 7-34. The highlight clipping display in Curve;·.
Figure 7-33. Before and after: A blue color cast corrected using the eyedroppers in Curves.
""eJ ch07_Curves Eyedroppers.jpg 1.
Add a Cu rves adjustment layer. Click the Show Cl ipping checkbox. To show the highlight cli pping, move the highlight control point (u pper-right corner) horizontally along the top of the grid to the left. Keep moving the point to the left until you begin to see cli pped areas (Figu re 7-34) . Using this method, the clipping display remains visible even i f you are not actively moving the control poi nt. (You can also Option/Air-click the h ighlight or shadow sliders under
With the add ition of a clipping display option to the Cu rves dialog box in Photoshop CS3, combi ned with the fact that it remains d isplayed even without having to resort to keyboard shortcut acrobatics, we feel there is less reason to perform endpoint neutralizing in Levels. After setting the highlight and shadow poi nts, we typically move on to Cu rves for additional a d j ustments to the midtones, so it makes sense to take care o f the entire process in C u rves from the begi nni ng.
Global Color Correction I
241
Using the gray point eyedropper. When using the gray point eyedroppers in either Levels or Cu rves to neutralize an image it is best used on a reas that should be a middle value neutral, colorless tone. Once you identify an area i n t h e image t h a t y o u think m ight w o r k for this purpose, b e sure t o zoom in close and double-check that there a r e no other color traces present that could confuse the gray poin t eyedropper a n d skew t h e results i n unexpected ways. I n the image seen in Figure 7-37, the power turbines are really gray and using the gray point eyedropper quickly alleviates the undesi red color cast.
Figure 7-35. The shadow clipping display in Curves
R: � G: B:
34/ 49 35/ 50 44/ 53
8-bU
8-bil 771 1022
Figure 7-36. The corrected shadow and highlight areas after they have been neutralized with the eyedropper tools in CUl·ves. Levels: Correcting m idtone color casts
If you neutralize the h ighlights and shadows and there still seems to be a color cast i n your image, then you need to address it by adj usting the mid tone slider for the different color channels in Levels. This can also be done in Curves and we'll cover that shortly. There a re a few ways to approach this adjustment, and the method you choose will depend on the image you're working with. The fi rst two methods work well for an image where there is an obvious midtone area that should be neutral; the first technique using the mid tone gray Eyedropper tool, and the second providing precise control via manual adj ustments.
242
I Chapter 7 : Color Correction
Figure 7-37. Before and After: In images with wefl-defined neutral areas, such as this view ofthe power plant at Hoover Dam, the gray point eyedropper worl(S quite well. Ii) Sf)
-1EJ ch07_GrayEyedropper.j pg 1.
Add a Levels or Curves adjustment layer, or double click a n existing Levels layer that you a re using to apply global color correction. Identify an a rea in the i mage that represents a middle tone that should be neutral for example, concrete, metal, gray clouds, or i n t h i s exam ple the grey metal t urbines.
To track the changes that this correction will apply, shift-click on the a rea to place a color sampler point there. This step is not strictly necessary for this technique, but if you want to see n umeric confirmation of the changes, i t can be helpfu l .
2.
I n t h e Levels dialog box, double-click t h e gray point eyedropper. I n the Color Picker you can see that the default target values for the eyedropper are 1 28, 1 28, 1 2 8-a middle gray (Zone V). Leave the values set to this default, and then click O K to close the Color Picker.
3.
4.
Click on the a rea you've identified as one that should be neutral. If you have placed a color sampler point, make sure the Caps Lock key is act i vated to display the precise crosshair cu rsor. Zoom i n for a close view and al ign the cursor with the color sampler point, and then click once to apply the target values of the gray point eyedropper as seen in Figure 7-38. I f the a rea you clicked on was originally truly neutral, the color cast will be removed from the mid tones. Check the values for the sampler point i n the I n fo palette. The values should be neutral (or very close to neutral), but they will not be 1 28, 1 28, 1 28. This tool tries to balance the color so that the targeted tone is neutral but it does not a ffect the overall brightness level of the tone. (Figure 7-39).
5.
If the color seems worse after clicking with the gray point eyed ropper, this indicates that the tone you clicked really wasn't a true neutral, and a corrective adj ustment has been made that introduces a d i fferent color cast. If this happens, try to fi nd a d i fferent a rea and use the gray point eyedropper again.
6.
I ",
Na.vl atar
Histo ram
4 1 / 29 18/ 18 1 1/ 18
R: " G: B:
8-bit +.
Info c:
.1. M : Y: K:
8-bit 1037 1012
X: Y:
114/ 98 97/ 98 82/ 98
B:
x
49/ 77/ 78/ 73/
61% 72% 67% 78%
I W r::;
H:
Figure 7-39. After clid(ing with the gray eyedropper the taJgeted midtones have been lleutralized.
If yo u i ntend to target a m etal l i c area that m i ght have
some
reflect i o n s
on
it
the
gray
point
eye d ropper may cau se u n expected resu lts as the reflected tones co u l d h ave col o rs in them that are n ot represe ntative of a true n e utral tone.
Manually neutralizing a midtone color cast. You can also
correct a mid tone color by placing a color sampler point" onto an area that should be neutral and manually adj usting the midtone sliders for the individual color channels. The I n fo palette shows which chan nels need to be adj usted. Here's how it works to create the results seen i n Figure 7-40.
le.ve.ls
1 ----,
G, ,---, ,,-, : Chao",': '",
� � � � Output levels: .
IOl
6
flliI
�
[yj Prevlew
Figure 7-38. The midtone gray point eyedropper in the Levels and Curves dialog box will neutralize the tone you elid( but /lot affect the brightness.
Figure 7-40. Three Orbs: Before and After versions.
.IIJ
Global Color Correction I
243
--'eJ ch07 M an u a l MidtoneFix j p g _
1.
2.
3.
4.
_
.
Identify the a rea i n t h e i mage that y o u feel should be neutral (or that you want to make neutral). I f you a re placing the poi n t prior to entering Levels or Curves, make the eyed ropper the active tool, and then Shift c lick. If you a re a l ready i n the Levels or Curves d i alog box, s h i ft-click to place a color sampler point on a neutral area. I n the night i mage of the three glowing orbs, we chose one of the l arge stones near the center steps. Take a look at the val ues i n the I n fo palette. Remember that in an RGB file a neutral value is equal amounts of red, green, and blue. By seeing which color values are h igher or lower, you can get a clue as to where you need to go to make the most of your adj ustment. For example, if the blue value is lagging behind both the red and the green v alues, as is the case with this i mage, it means there is a yellow color cast in the mid tones (red and green combine to create yellow, and yellow is the opposite of blue). Add a Levels adj ustment layer and open the Channel menu at the top of the dialog box, and choose the bl ue channel to begin your adj ustments. You should start your adj ustments on the channel that has the most deviation from the other two. For this example, we need to remove a yellow color cast, so the blue channel is the place to begi n . In the blue channel, move the middle slider to the left to add more blue to the mid tones. Keep an eye on the adj usted values in the I n fo palette (the val ues on the right side of the slash). The aim is to a d j ust one or more of the channels until you arrive at a neutral tone (signified by equal amounts of red, green, and blue) i n the I n fo palette. For the photo of the th ree orbs, we increased the blue mid tone value and decreased the red m i d tone value until the values for the color sampler point were equal (Figu re 7-4 1).
F o r f i n e adj u stm e nts, if y o u h i g h l ight the n u meric val u e in the m i dtone box, you can use the u p or down arrow keys to move the n u m be rs i n one point i n crements. To adj u st them by ten poi nts at a ti me, use the S h ift key with the arrow keys.
244
I Chapter 7: Color Correction
...
K:
8-blt
C: M: Y: K: a-bit
...
Figure 7-41. To neutralize the stone wall and remove the color alst from the image, " ed was subtracted and blue was added by adjusting the midtone sliderfor those channels in a Levels adjustment laYe!:
To ident ify the mathem atical n e u t ral val u e, place the Eye d ropper over the m i dtone area yo u want to target to d i s p l ay the RGB val ues i n the I nfo pal ette. Add the th ree RG B m i d p o i n t va l u es and d ivide by th ree to dete r m i n e the average. T h i s va l u e can se rve as the target p o i n t w h e n adj u sti ng m i dtones color m a n u a l ly.
While these techniques are good for ensuring that a mid tone value is numerically neutral, keep in mind that numbers alone should not rule over your image. What counts is how it looks to you and whether you feel that the tonal and color balance is right for the image. We frequently use these steps to neutralize values in a photograph only to manually move them away from neutral to enhance the emotional and subjective value of an image. Not all color casts are i nherently "bad," and life, perception, and memory are rarely neutral. Know how to identi fy these characteristics in your images and be familiar with the techniques for neutralizing the highlights, shadows, and mid tones, but don't let by-the-numbers neutrality dictate what your images should look l ike. While the presence of a neutral tone, or a tone that should be neutral, can be a very useful guide i n determining the cause of a color cast and correcting it, not all images have tonal values that should be neutral. If you find yourself working on such an image and a color cast is present, you will have to manually adjust the Levels midtone sliders for each of the three color channels until you shift the color balance more to your liking. Correcting a color cast with n o neutral guide.
Although there may be no neutral guide, you can make some preliminary determinations about the color cast that will help you decide which color adj ustments a re li kely to be the most effective. Fi rst, try to qualify the color cast as warm or cool . Warm color casts are typically dominated by yellow, red, or magenta. Cooler casts are created by the presence of blue, cyan, or green. A color cast can be created by primarily one color, or it may be the result of two colors. For example, blue and green combine to make cyan. Some casts also can be a combi nation of warm and cool, such as a yellow-green cast, or a blue-magenta cast. Once you have narrowed down what colors might be causing the color cast, knowing the basic structure of the color wheel, complementa ry (opposite) colors and component colors, is very useful for determining which color channels to adjust (Figures 7-42 , 7-43, a n d 7-44). For exam ple, i f you determine that a n image is suffering from a cyan color cast, the red chan nel would be the most l ikely place to try to fix it si nce red is the opposite of cyan. The presence of a cyan cast indicates that green and blue (the building blocks of cyan) are too domina nt, and a little red is needed to lessen their i n fl uence. Of course, you coul d also reduce green and blue, which would cause an increase of magenta and yellow, respectively. And, i f you're familiar with the structure of the color wheel, you know that magenta and yellow combine to make red, so it ends up being the same thing!
I
Channel:
! Blue
I
:JI
Input levels :
Channel:
( Blue
:Ij
Input Levels:
, .=] �. +-l [Ol
[l:OOl
Illil
[Ol
Output Levels:
.� ro-I
[l:OOl
Illil
�
Figure 7-44. The midtone sliderfor the Blue input levels adds blue or yellow to the midtones. The input and output shadow and highlight slidersfor the blue channel allow you to add blue or yellow to the shadows or highlights.
I n the example photograph of a stained glass w indow on a luxury rail road car seen i n Figure 7-45, the warm, yellow color cast is caused by i n terior d isplay l ighting and is typical for that type of ill umination. No areas of the photo should be neutral, so any color correction is based on knowledge of the color relationships (that is, k nowing what colors cause a cast and how to counteract them), as well as how you want the i mage to look.
Figure 7-42. The midtone sliderfor the Red input levels adds red or its opposite colol; cyan, to the midtones. The input and output shadow and highlight slidersfor the red channel allow you to add red or cyan to the shadows or highlights. Channel:
Grun
I
Channel:
! Green
:11
Figure 7-43. The midtone sliderfor the Green input levels adds green or magenta to the midtones. The input and output shadow and highlight slidersfor the green channel allow you to add green or magenta to the shadows or highlightJ.
Figure 7-45. Top: An image with no neutral guides �·ujferingfr0171 a strong reddish yellow color cast. Bottom: the corrected version. Sf)
Global Color Correction I
245
JEJ ch07_N o N eutraIG u ide . j pg 1.
The fi rst step in any color correction is to evaluate the image and determine what the problem is. For this image there is an obvious warm, yellow cast. Since yel low is the main problem color, you should fi rst try working with the color that represents the opposite of yellow, which is bl ue.
2.
Add a Levels adj ustment layer and select the Blue channel from the menu at the top of the dialog box. Add a generous dose of blue to the midtones by moving the m idtone slider to the left (Figure 7-46).
3.
Si nce red is also a component of the color cast, select the red channel and subtract a bit of red from both the mid tones and the shadows by moving those sliders to the right to add cyan (Figure 7-47). Si nce there is no neutral a rea to guide you, adjustments such as these are based on how you want the i mage to look. Channel:
: Blue
:
1 ------.
with the ability to precisely d i rect the e ffect th rough a layer mask, makes Cu rves a very effective color correction tool. As has been noted th roughout the book, anything that can be done with Levels can also be done with Cu rves, but Curves offers greater control and precision.
U nder th e Hoo d: Manual Color Correction with Curves To apply manual color correction, you need to work on the ' individ ual channels that are accessed in the channel menu at the top of the Cu rves d ialog box. The color relationshi ps that can be affected i n these individual chan nel curves a re the same as the opposites on the color w heel and the same a s the color relationships found i n the i ndiv idual channel controls in Levels. Tn an RGB image, adj ustments to the red curve will affect red and its opposite, cyan; modi fications to the green curve will a ffect green and its opposite, magenta; and changes to the blue curve will a ffect blue and its opposite, yellow (Figure 7-48 and Figure 7-49). Developing a solid understand i ng of these fundamental color relationships is a key ingred ient for perfo rm i ng good color correction work in Photoshop.
Figure 7-46. The Levels adjustments for the blue channel ofthe vintage train-car photo. : ! -----, Channel: �...._ d___I...'-' R. e_
Figure 7-48. The color relationships controlled by the red, green, and blue curves.
6
[2s5l Figure 7-47. The Levels adjustmentsfor the red channel ofthe vintage train car image.
The same modification to the i nd iv idual color channels could be applied with Curves, of course, and with a great deal more control over which a reas of the tonal range are affected. T n the following section, we'll take a closer look at using Curves for color correction.
The Powe r of Cu rves The Curves feature is one of the most powerful color control features in Photoshop because i t enables you to target and adjust specific brightness and color values. This, combined
246
I Chapter 7: Color Correction
Figure 7-49. By moving the endpoints in the Curves color channels, you can add red, green, blue, cyan, magenta, or yellow to the highlights or shadows ofan image. Cu rves exa m p l e 1 : Cleaning the pond
As in a lot ofdigital darkroom work, the fi rst step is to identify the problem and then decide how to add ress it. Identifying a colorcast is a skill you can develop and refine by looking at a lot of images. The advantage of looking at i mages in Photos hop is that you can use the I n fo palette to identify colorcasts that aren't obvious. At fi rst glance the water in Figure 7-50 has an overall m u rky, greenish cast that in our opinion is not very pleasing at all. There is reall y no obvious
neutral that we can use as a guide here; the reflection of the overcast sky on the water is as close to a neutral as we're likely to fi nd, but it will d o nothing to help the color of the water. And al though the depths of the water are probably not neutral, we can still use them to hel p us zero in on what might be causing the color cast.
Figure 7-51. By sampling an area that should be neutral, 01· reasonably close to neutral, the Info palette can 11e1p reveal the cause of the color cast. In this case, the lack of blue is creating a yellow color cast. 2.
Add a Curves adjustment layer to the image. Find an a rea of dark water to use as a reference point. I n this example we used the dark a rea in the center of the image j ust to the left of the swishing tail (see Figure 75 1 ). Shift-click to add a color sampler point there. Note, if the r nfo palette is not al ready open, as soon as you add a sampler point Photoshop will open it for you.
3.
I n the Channel pull down menu at the top of the Curves dialog box, choose B l ue. Cmd/Ctrl-c1ick the reference color sampler point i n the image to pinpoint the exact location on the curve. With an eye on the val ues for the sampled point i n the I n fo palette, raise the curve to bring the n u meric blue value in line with the red and green val ues. The goal is not to try to make all val ues perfectly identical, but to increase the blue to cou nter the dominant yellow created by the higher val ues of red and green. For example, in the version of the image that we worked on and the point we sampled, the red, green, and blue val ues were 50, 47, and 2 5 , respectively. By increasing the blue until i ts value was at 42, the yellow-green color was removed from the water quite nicely (Figure 7-52).
Figure 7-50. In the original image ofthe I(oi, the water has a murRY, greenish-yellow color cast. The corrected version looks much bette!: \/
'1J ch07_C u rvesExa m p l e1 .jpg 1.
Look at the I n fo palette (Window--> I n fo) and move the cu rsor over a dark area of the wate r. The numbers reveal that red and green are nearly the same in value, and blue is lagging behind (Figure 7-5 1 ). This tells us that red and green are stronger and more dominant in the image. The result of com bining red and green is yellow, and as a n inte resting coincidence, yel low happens to be the opposite of blue. So green is not really the offending color here, though it does play a supporting role. Yellow is the primary color cast a ffecting the water, which ma kes the water look less blue and less i nviting and subjectively stated - murky and muddy.
:1
vi ato R: J( G :
B:
iSl
rom
117/ 117 B7/ B7 SO/ BO
a-bit +.
X: Y:
HIR: J( G :
B:
Info
c: ", 1'.1 : Y: K: a-bit
x
27/ 65/ B7/ 15/
-=
32% 67% 54% 12%
1490 952 SO/ 5 0 47/ 47 25/ 42
Doc: 6.75"1/6.75M
. input: I36 I
Show
Clipping
Figure 7-52. To reduce the strength ofthe yellow cast, the blue curve is raised ulltil the blue value in the Info palette becomes closer to the red and green numbers. 4.
While this adj ustment works well for the water, it tends to make the more neutral reflection of the overcast sky look a bit too blue (Figure 7-53). Fixing that is fairly straightforward since Cu rves provides so
The Power of Curves I
247
much control over where along the tonal range a color adj ustment is applied. Click in the image on one of those l ighter reflective a reas to determine where on the c u rve that tone is located.
6.
To fu rther suggest the cool temperature of the water, we also chose to access the red chan nel c u rv e and subtract a small amount of red from the darker a reas in the image (from input level 45 to output level 42see Figure 7-5 5). This works well to c reate a color that is plausible for a koi poncl . The main thing here is not to be obvious. You a re not really adding cyan so much as removing a bit of red .
7.
To strengthen the image and give it a solid visual foundation, we decided to increase the contrast by darkening the shadows and carefully l ightening the highl ights. Return to the composite RGB channel and increase the contrast by darkening the shadow a reas and slightly raising the upper-middle part o f the c u rve (Figu re 7-56).
:J
Figul'e 7-53. The initial adjustment to the blue curve cause>' the highlights in the water reflection to become too blue. 5.
Place a point on the blue curve at that location and d rag down towards the original baseline of the curve to prevent that tonal region from being a ffected by the bl ue increase in the darker areas of the water. For finer cont rol tap the down a rrow key to lower the highl ights in the blue channel in one level i ncrements (Figure 7-54).
Figure 7-55. Tile adjuJt1nent made to the shadow areas a/the l'eel curve.
�: 131 I Show Cli pping
Figure 7-56. Thefinal adjustment made to the overall RCB curve to darl(ell the water and increase contrast. Cu rves exa m p l e 2 : Backyard witch
Figure 7-54. This modification to the blue curve prevel1ls the highlightsfrom becoming too blue.
248
I Chapter 7: Color Correction
The environment that a subject is in often infl uences the light illuminating a subject. For example, a brightly colored wall can kick in color reflections onto a person or as seen in Figure 7-57 in the photograph of a girl playing d ress-up as a witch, there is an unpleasant cyan/green color cast from the surrounding envi ronment. This is cause I by the fact that there is so much bright green vegetation in the shot, it is completely in the shade, and the late a fternoon light is being heavily filtered by the canopy of green leaves in the trees above the
scene. To make matters more challenging, there are no areas that might serve as a neutral guide in this photo. The house in the background is a m uted green color, and the trim is not a neutral hue. The black of the d ress actually has some blue in it, and using this a rea as a neutral guide would add too much yellow to the image. With no reliable neutrals to guide us, we have to proceed based on ou r initial visual evaluation. If you know the basic color relationships, however, then the recipe for fixing this color cast is clear. If too much cyan and green are present, then we must subtract those colors by adding i n their opposites, which a r e red and magenta.
2.
A good way to apply extremely subtle adj ustments such as this one is to overdo it and then slowly back away from that point. Move the position of the point upward until you can see that red is obviously being added. Now use the down arrow key to back it off slightly until it looks good. Our in put value was 1 28, and the output value was 1 33 (Figure 7-58). The goal is to remove the obvious cyan cast without add i ng an obvious red cast.
3.
Since there is so m uch green in this i mage, i t makes sense to see how subtracting green will a ffect the i mage. Choose G reen from the channel d rop-down menu. As before, click i n the center of the grid to place a control point there (level 1 28). Press the Shift key and tap the down a rrow once. The use of the Shift key moves the poin t in 1 0-level increments . If you can't see much d i fference, then Shift-tap the down arrow once more to move it down 20 levels ( 1 28 to 1 08). This is obviously too m uch, so now you can begin to reduce the magenta in I -level increments until it is not apparent any more. F rom a starting input value of 1 28, our final green adj ustment was an output value of 1 20 (Figu re 7-58). :1
� 1128 I 1/1/1/1
o Show Clippln9
Figure 7-58. The red and green curves adjustmentsfor the Bac/fyard Witch.
Auto Color Correction Options
Figure 7-57. A predominantly green loeation, coupled with shade and late afternoon lightjiltered by a eanopy ofgreen leaves has created a cyan green color east in this image. In the corrected version, subtle adju.itments to the red and green curves have removed the color east.
-t ch07_C u rvesExa m p l e2.jpg 1.
Add a Curves adj ustment layer, and i n the dialog box, open the Channel menu and choose Red. Si nce this is an overall color cast and not specific to a particular tonal region, place a poi nt right in the middle of the cu rve grid (level 1 28). Only a very slight adj ustment is needed here, so use the up a rrow key on the keyboard. Each tap of the arrow key w i l l raise the control point by a si ngle increment.
In your travels th rough the Photoshop i n terface, in the Image->Adj ustments menu you may have seen Auto Levels, Auto Contrast, and Auto Color. While the promise of Photoshop's magic elves applying color and tone correction for you may sound enticing, we generally do not favor relying on the auto corrections simply because, depending on the color and tonal structure of the i mages you use them on, the final results are very hard to predict. Admittedly, for some images, the auto corrections can provide a j um pstart to get you headed in the right d i rection. But rather than access these commands from the I mage->Adjustments menu where you have no control over them, you can access them in both the Levels and Curves dialog box, which a llows you to apply them as an adj ustment layer and fi ne-tune the results with the regular Levels or Cu rves controls (Figure 7-59).
The Power of Cu rves I
249
Levels
Channel:
!,-"R"" GB,---, -- ,,-, : I ---�
Output Levels:
.
101 OutPUf:
....
.;:;;;� ;;;;:;;: "
l;t I;'1./I 0 Show Clipping G Curve Display Options -----Input:
Figure 7-59. The best way to access the Auto Color Corrections is by pressing the Options button in Levels or CUl·ves.
Figure 7-60. This scan Fain an older slide is low contrast and has an obvious yellow c% r cast. i( ) Sf) Enhance Monochromatic Contrast (Auto Contrast)
Auto corrections deconstructed
To understand when the auto correction options might be worth while, it helps to understand what's going on under the hood . To try these out, add a Levels or Curves adj ustment layer to a n image (the auto options a re identical). Click the Options button to bring up the Auto Color Correction Options d ialog box. The best thing about applying the auto options via an adj ustment layer is that you can test-d rive each one to see if it does any good, and when you click OK, you are still i n the Levels or Cu rves d ialog box where additional fine-tuning can be applied. Algorithms. The three auto correction algorithms are exactly the same as the three auto options i n the I mage� Adjustments menu. Enhance Monochromatic Contrast is equivalent to Auto Contrast; Enhance Per Channel Contrast is the same as Auto Levels; and Find Light and Dark Colors is the same as Auto Color. Plus there is also an additional option called Snap Neutral Midtones. To demonstrate how each of these affects a n image, we've chosen a scan from a slide taken in New Mexico that has a definite color cast as well as flat contrast (Figu re 7-60) .
This algorithm will increase contrast in the photo by applying the same endpoint adj ustment for each color channel (Figure 7-6 1 ). While this choice can someti mes work for a quick contrast adj ustment, it does not add ress any color cast issues, which makes it the least useful of the three options. Enhance Per Channel Contrast (Auto Levels)
This algorithm will apply d i fferent adj ustments to each color channel with the goal of i ncreasing contrast and fixing color casts (Figure 7-62). This is somewhat more useful, but the results a re very i mage-dependent; sometimes i t works wel l , but j ust as often i t is l i kely to make the i m age look worse. Find Dark and Light Colors (Auto Color)
This is the most promising of the th ree algorithms. I t fi n d s t h e darkest a n d brightest tones i n t h e i m age a n d adj usts t h e m t o b e a neutral h ighlight a n d shadow using the target values for the Levels or Cu rves eyedroppers as a guide (Figure 7-63). When the Snap Neutral Midtones is checked, it locates any midtone value that is close to neutral and adj usts it so it becomes neutral. For images with a noticeable color cast, Enhance Per Channel Contrast and Find Dark and Light Colors work the best, and the result is often (but not al ways) improved if you turn on Snap Neutral Midtones.
250
I Chapter 7 : Color Correction
Target colors and clipping. The default shadow and
highlight cli pping values a re 0.1 0%. This means that the auto corrections will i ntroduce this amount of cl ipping (forcing tones to either total black or wh ite) into those endpoint regions. In general, the clipping defaults are a good starting point, and they can hel p to im prove the contrast of images, but if after using this feature you measure the bright highlights in your image and feel they a re too bright, you should reduce the clipping setting. To influence the overall color balance, click the Midtones color swatch and adjust the values for R, G, or B in the Color Picker to change the color balance to something different (it's best to make these adjustments i n small increments). For example, if you wanted an overall warmer feel to images, you could use a midtone value that contained more red and yellow. Figure 7-61. Enhance Monochromatic Contrast: The contrast is certainly improved, but the color cast remains.
Auto cor rectio ns: The botto m l i n e
Until y o u develop some understanding of w h a t i mages are good candidates for this feature and how the auto corrections a re evaluating and then adjusting a photo, the success rate may seem very arbitra ry. Every image is d i fferent, with its own individ ual tonal and color matrix and, as the saying goes, your m ileage may v ary. If you a re a l ready comfortable with making adj ustments i n Levels or Curves, then you'll probably feel the auto corrections are too un predictable. I f you're new to color correction, however, then these might get you started i f you're not sure how to proceed with a tricky image. The most i m portant thing is that you should not rely on just these settings to make everything perfect. If you do use the auto corrections, they should be viewed as a starting point. Additional fi ne-tuning should always be applied with the Levels or Curves controls. Figure 7-62. Enhance Per Channel Contrast: The auto correction worl(S very well on this image, though it is perhapsjust a bit too blue. Al9orilhmJ O [n�n(tlolonO(h'omllk COnlrllt
"hi r'p " nnsICpn!.all ennd O"k & Ught Colon
�
I�
�
Co l o r E n h a n ce m e nts Once you have corrected the color in an image, the next step is to enhance the colors. This can be done to either the overall color palette of the i m age, such as by i ncreasing or decreasing color saturation, or by changing the character or component val ues of a specific range of colors. There are many ways to enhance color i n Photoshop, as well as some interesting features in Adobe Camera Raw and Photoshop Lightroom. Some a re fai rl y general while others offer more specific and detailed controls. In this section we'll explore some of the most useful color e n hancement tools to make your images look thei r very best.
Enh ancing Col or with Hue/ Saturation
Figure 7-63. Find Darl( and Light Colors: This auto correction is a/so 110t bad, though it is more obviously too blue. (Note how cool blue the adobe wall is.)
The H ue/Satu ration command enables you to change the "flavor" of colors (the h ue) by addi ng or subtracting their component colors, as well as i ncreasing or decreasing the i ntensity (saturation) and l u m i nosity (lightness) of colors. All colors in the i mage can be a ffected when using the default
Color Enhancements I
251
Master controls or you can access specific color groups by opening the Edit menu at the top of the dialog box. As with all such color and tonal modifications, the best way to apply Hue/Saturation is by using a n adj ustment layer.
Edit:
; Master
Hue/ Saturation
� :l
Hue:
�
rol
Saturation:
�
H ue/Satu ration controls
I n Master mode, moving the Hue slider will cause all the color values in the image to s h i ft d ramatically (Figure 7-64). I f you look at the color spectrum bars at the bottom of the dialog box in Figure 7-65, you'll see how the colors in the lower bar a re changing i n relation to those i n the upper bar (think of these spectrum bars as a horizontal representation of the color wheel). The upper spectrum bar represents the original colors, and the lower one shows how they a re changing.
I-10oI
Lightness:
OK Cancel
) )
load ... Save ...
�
0 Colorize
,---,
L� .! £ � Preview �
'''-
Figul'e 7-65. The highlighted areas on the spectrum bars show how the coloH ofblue and orange in the Holiday Motel image have shifted to green and magenta.
Figure 7-64. The radical change to the Holiday Motel photo was achieved by adjusting the Hue value to 100 ill the Hue/Saturation dialog box. so -
The Saturation slider increases or decreases the saturation (i ntensity) of the colors (Figure 7-66). I t's easy to go over the top with this by adding too much saturation. Negative satu ration values result in a more m u ted color palette, and moving the slider all the way ta the left (- 1 00) will remove all color val ues from the phato and create a grayscale effect.
252
I Chapter 7: Color Correction
Figure 7-66. Modifications to saturation: The center image is the original. The one 0/1 the top has been desaturated by -25. [/1 the version 012 the bottom, the color saturation has been increased to +25. S/ )
Creative uses for the Lightness slider Wh i l e tota l l y desatu rat i n g an i m age does produce a grayscale
i nterpretation that can
sometimes
work we l l, we d o not favo r u s i n g this m ethod for a b l ack-and-wh ite conve rsion because of the lack of i nterpretive opti o n s ( F i g u re 7-67) . In add ition to th is, the b rightness relat i o n s h i ps between very bright colors sometimes do not tra n s l ate we l l when th is tech n i q u e i s u se d . See Chapter 5 for more effective tech n i q u es o n translating color i n to b l ack a n d w h i te.
Though we feel that the Lightness slider is not very useful for most images when a pplied to the entire image, there can be exceptions to this. Using positive values, for example, is very effective for creating a ghosted-back look that is often employed for overlaying type on top of images, or collaging photos together where one image is clear and d istinct and the other image serves as a faded background. In Figure 769, a H ue/Saturation layer with the Lightness set to + 60 was used to fade back the image of the palace windows so the D resden type layer is more visible. A layer m ask was added to the Hue/Satu ration layer with the silhouette of a famous Dresden church in gray to partially mask the l ightening effect on the wi ndows layer.
Figure 7-67. The version on the left was created by totally desalt/rating the colon' in Hue/Saturation. The one on the right was created using the Blac/( & White adjustment layel; which offers much greater control over how the individual colors are translated into gray tone>'.
The Lightness slider is the least useful of the th ree controls, especially for adjustments that a ffect all the colors, though it does offer possibil ities for fine-tuning individual colors. The main reason i t is not that effecti ve for the overal l ad j ustment is because it alters the brightness values i n a way that interferes with the natural l um i nance and contrast relationships i n the image ( Figure 7-68). Adj usting the Lightness to positive values creates the effect of a translucent white overlay. Using negative values with the Lightness slider will darken the image, but not in a way that looks natu ral or convincing. If you want to brighten or darken an image, use Levels or Curves for both better e ffects a n d greater control .
Figure 7-69. Using the Lightness >'Iider with positive values can create a tramlucel1t white overlay that worl(S wellfor some collage effects or for placing text over an image. 1/)
Figure 7-68. The effects ofthe Lightnes>' slider: The center image is the original photo. in the version 017 the left, the Lightness has been set to -40. /n the version all the right the Lightness has been set to + 40. III
Color Enhancements I
253
For some images, especially black-and-white photos, applying significant darkening with the Lightness slider can create the look of an antique photo process (that is, dark and low contrast) or a dark, moody atmosphere (Figure 7-70). As with all things creative, the best way to find unexpected uses for the controls is to experiment to see if using them a certain way creates an effect that works well for a certain image or project.
Figure 7-71. The Byodo-ln Buddhist temple on the island ofO'ahu.
, Sf)
-IEJ ch07_Tem p le.jpg
Figure 7-70. In this image, the top Hue/Saturation layer darkens the photo and lowers contrast with a lightness setting of -65. A layer mas/( is used to lessen the effect on the woman's hail: 1
From 75° to 105° the effect is gradually feathered to zero
Click and drag on the feather "handles" to change the size of the feather zone
Figure 7-81. The color spectrum bars in the Hue/Saturation dialog box display information about the range ofcolor being adjusted and also enable you to customize affected colors.
Figure 7-82. The degree numbers at the bottom ofthe Hue/Saturation dialog box show how the effect isfeathered in relation to the position of colors on ti,e color w/leel.
Color En h a ncements I
257
One possible way around this is to use the Vibrance slider in Camera Raw or Photoshop Lightroom. Vibrance is a close cousin of the Saturation slider but its changes are applied more delicately. The saturation of l ess-saturated colors is i ncreased with l ess impact on the higher-saturated colors, a nd clipping is m i nimized as colors approach ful l saturation. I n the i mage of th e toy microscope set (Figure 7-83), i ncreasing the saturation a ffects all the colors (Figure 7-84). The red item to the right of the mic roscope, al ready a bright red, becomes ev en more saturated. The Vibrance slider in Camera Raw i ncreases the saturation of the duller y el lows, oranges a nd blues, but does not affect the bright red that al ready has a high l ev el of saturation. For portrait work, Vibrance is also a better a lternative than Saturation because it prev ents skin tones from becoming oversaturated (Figure 7-85). Figure 7-85. The Vibrance slider in Camera Raw (above) and Lightroom increases the saturation oj-more muted colors while leaving higher saturated colors alone.
Photo Filter Adjustment layers The Photo Filter feature (Lay er->New Adjustment Layer->Photo Filter) is d esigned to mimic th e properties of traditional warm i ng and cooli ng filters that are used on lenses to mod ify the color temperature of the l ight in a scene. The fi rst six choices i n the fi lter m enu in the dialog box are digital v ersions of traditional fil m-based col or balancing filters that have been used for a long time. These can som etim es be useful for subtle adj ustments to the overall color balance of an i mage, especially i f i t j ust needs a littl e warming or cooli ng. The remaini ng choices cover a range of d i fferent colors (Figure 7-86). Figure 7-83. The original image with no adjustments.
Red Orange Yellow Green Cyan Blue Violet Magenta Sepia Deep Red Deep Blue Deep Emerald Deep Yellow Underwater
Figure 7-84. The Saturation controls in Photoshop, Camera Raw (above), and Lightroom increase saturation levelsfor all colors in the Image.
258
I Chapter 7: Color Correction
Figure 7-86. The Photo Filter dialog box showing the availablefilter choices.
The warming a nd cooling filters are the most useful and the ones we tend to use most often. The 8 5 and LBA (warming) and 80 and L B B (cooli ng) are color correction filters i ntended for applying minor fi ne-tu ni ng to a n image's white balance. These can be useful for i mages with a mild color cast caused by i nterior tungsten l ighting, or images that are a bit too cool , such as those photographed u nd er shady conditions. The 8 1 warming and 82 cooli ng fil ter create similar effects but are a bit stronger. In the photo taken i nside a World Wa r I I - era submarine (Figure 7-87) , the typical yellow color cast caused by the tungsten l ight sources is reduced by using the Cool i ng Filter (82) set to a d ensity of35%. Sinc e this image has so many a reas that could serve as neutral guid es, this color cast could also be corrected i n Levels or Curve using the eyedropper m ethod shown ea rli er. For additiona l i nformation on Photo Filters please see the section on Lab color at the end of this chapter as w el l as Chapter 8 , "Creativ e Color".
c a n be useful as a q u i c k fi x for J P EG files or older sca nned i mages. The violin photo is a scan from a slide and suffers from an ugly green color cast caused by the fluorescent l ighting in the d isplay case. By choosi ng a Viol et filter (an approximate com pl ement color for a greenish cast) with a d ensity of 35%, much of the u np l easant color cast has been neutralized (Figure 7-88). The color balanc e is not perfect, but it's a lot better than the original. We'll revisit this image in the next exercise to explore more precise ways to im prove the color balance.
ocotor:.
Density:
9
[35J"
:l� � �Prevl�w
Figure 7-88. The Violet Photo Filter set to 35% Demity does battle with a nasty green cast caused byfluorescent ligllting. \/)
Figure 7-87. The Cooling Filter (82) at 35% Density proved effective at clearing up the color cast caused by the lighting ilwde the submari17e. (,) Sf)
There's also a list of other choices based on specific colors, or you can choose the color option and specify you r own custom color. Either the preset color choices or a custom color can be used to neutralize a color cast by sel ecti ng a complementary color. W h i l e this is not our first choice for removi ng a color cast (that would be Curves or, if a raw fil e, the conversion controls i n Camera Raw or Lightroom), i t
local Color Correction
After you have applied adj ustments to a ffect the global (overall) color in a photograph, the next step is to focus on corrections that are appli ed to specific a r eas. As with local mod i fi cations to tonal qualities of an i mage (that is, lumina nce a nd contrast), the use of layer masks will gi v e you the greatest control and fl exibility over how the local color corrections are appli ed .
Loca l Color Correction I
259
Local Corrections with Curves Tn addition to its prowess as an overall color-correction tool , when used with a layer mask, Curves is also ideal for applying local corrections to specific areas of an image. The i m portant thing to understand about local corrections with Curves is that the only d i fference from overall corrections is that they are precisely targeted to a ffect certain a reas. The basic m ethodology of evaluation, id entify i ng the color issues that need correcting a nd d etermini ng the color recipe to use, is essentially the sam e that you would use for overall corrections. To explore one example of using Curves on a specific portion of a photo, we'll continue with the viol i n photograph that we used t o demonstrate the Photo Filter adj ustm ent layer (Figure 7-89).
2.
Use the Ey edropper tool to read the values in the tailpiece (the part of the viol i n on the body where the strings are attached). This is a n a r ea that can serve as a neutral guide and provi d e som e clue as to what is causing the l i nger i ng color cast (Figure 7-90). In most of this area, green is the higher v al u e, which makes sense for a color cast caused by fluorescent l ight (this is typically a mixture of green and cyan). Shift-cl ick with the eyed ropper to place a color sampler point there.
3.
Use the Lasso tool with a feather value of zero to make a rough selection of the body of the violin (Figure 79 1 ). Don't worry about the neck of the i nstrument and add a Curves Adj ustm ent layer.
75/ 75 111/ 111 99/ 99
Figure 7-90. The tailpiece of the violin serves as a good neutral guide. The valuesfrom the color sampler point show that green is still causing problems.
Figure 7-89. The antique violin, before and afto:
"'eJ ch7_vi o l in.j pg
Local color correction with layer masks 1.
I f you have not al ready done so, add a Photo Filter adj ustment layer to the v ioli n photo. From the Filter menu, choose Violet, a nd then set the Density to 35% with Preserve Luminosity checked . Click OK. This creates a d ecent overall color balance, but the viol i n is still too cool and bears sl ight traces of the original color cast. Figure 7-91. A rough selection ofthe violin body becomes a layer mas/( when a Curves adjustment layer is added.
260
I Chapter 7: ColO!" Correction
4.
I n the Cu rves dialog box, go to the green channel . Click the body of the v iol i n i n t h e image to see where on the curve you need to make the adj ustm ent. Since the i nstrument is primarily darker tones, this will be the lower part of the curve. Cl ick the curve to place a poi nt there and begin to lower it sl ightly to subtract green and add magenta (Figu re 7-92). Remember that you can also tap the down a rrow k ey to make fine adj ustments to the active control point.
5.
Switch to the blue curve a nd lower the curve to add in a bit of yellow (Figure 7-93). This lowers the blue value of the color sampler point on the tailpiece and also hel ps to warm up the violin to c reate a more plausible color for a fi ne wood en i nstrument.
,utpur 72
Fine-tuning the layer mask
7.
The final steps are to fine-tune the layer mask for th e viol i n. Use the Polygonal Lasso tool to make a sel ection of the violin's neck. Press D to set the foreground colors and then X to place white i nto th e background swatch . With the Cur v es layer mask acti ve, press the Del ete k ey to fi ll the sel ection with white. This adds the color correction to the neck of the i nstrum ent. Choose Sel ect-Desel ect.
8.
Choose Filter-Blur-Gaussian Blur to feather the edges of the layer mask. The blur radius that you use is always influenced by the resolution of the image; a photo with more pixels in it can use a higher radius val ue. For this image, a radius of5 pixels works well. Cl ick OK.
9.
Depend i ng on how accurate your origi nal Lasso sel ection was, you m ight need to touch up som e of the edges of the layer mask with the B rush tool. Double click the Zoom tool to zoom in to a 100% vi ew, a nd scroll a round the image to check the edges. Use a soft edged brush tip of an a ppropriate size to perform any mask touch-ups (Figure 7-94). Pai nting with white will show the color correction for the violin; paint i ng with black will mask it.
•
� 1 98 I
1/1/1/1 o Show Clipping
Figure 7-92. The green curvefor the violin.
Channel: ! 81ue
I:j
�
-�
,utpur 102
•
nput:
I
113
I j
C>
I 1/1/1/1 Show Clipping
Figure 7-93. The blue curvefor the violin. 6.
You can also experiment with sl ightly i ncreasi ng the red curve, but w e fel t this added too much red. Even though the values for the sampl e point on the tailpi ece a re not equal, our aim here is not neutrality of that item, but making the overall viol i n look good. Click OK to apply the Curve adj ustm ent.
Figure 7-94. Thefin al layer maskfor the violin after adding in the neck and the bow.
[n addition to i l l ustrating local color correction with Cu rves, this example also shows that you don't necessa rily have to have a super precise sel ection or mask to apply changes to only a portion of the i mage. True, i n some cases (such as collage work), a precise mask may be required, and for a shape such as this, th e Pen tool would be the obv ious choice for creating the i nitial sel ection. But for changes to color and tone, sta rting out with a looser sel ection often works fine, especially if you suspect that the mask edges may end up on the recei v i ng end of the Gaussian B l u r fil ter.
Loca l Color Correction I
261
Color fill layers to neutralize color casts There's one last thing about the v iolin image that needs add ressing: The wall has a n uneven color balance that looks a bit strange. To fix this, we'll take a quick side trip away from Cu rves and use an adj ustment layer not normally noted for its color correction properties. 1.
Hold down Option/Alt and click the Add Adj ustment Layer button at the bottom of the Layers palette. (The Option/Al t key needs to be down before you click the a d justment layer icon.) Choose Solid Color at the top of the list. The addition of the Option or Alt key forces the New Layer d i alog box to come up. From the Mode menu at the bottom of this dialog box, choose Color. Click OK.
2.
In the Color Picker, move the cursor all the way to the left of the large color selector box to choose a neutral gray (Figure 7-95). The brightness of the gray does not matter, as long as it is a gray. I f you want the wall to have a slight color flavor, such as a warm tone, you can choose a general h ue range from the spectrum bar and then move the selector circle in slightly to the right. Click OK.
New Layer
I
Figure 7-96. The initial effect ofthe neutral Color Fill layer will be a monochromatic version ofthe image.
4.
Choose Image-Adj ustments-I nvert to i nvert the layer mask, turning it black and hiding this grayscale effect. Press D to set the default colors, w h ich m akes white the foreground color. Select the B rush tool i n the Tool palette. In the Options bar, set the brush opacity to 50% from the Brush Picker on the far left of the Options bar; choose a large-sized brush that will let you cover the wall i n only a few strokes (for this i mage we used a 300-pixel brush).
5.
Brush over the wall to paint i n the neutra l tones created by the Color Fil l layer (Figure 7-97). Take care not to paint over the edges of the violin or any of the fabric that i t rests on. I f you do, just press X to exchange the colors and paint with black using a smaller brush tip to remove the neutral tones from any of the v iol i n edges.
Name: IColor Fill 1 I ( OK ) o Use Previous Layer to Create Clipping Mask ( Cancel) Color: ( DNone , :I ' :l rpacitY: � % Mode: [ Color
Figul·e 7-95. Top: The New Layer dialog boxfor the Color Fill layer with the mode set to ColOJ: Bottom: Selecting a neutral gray tone in the Color Picl(e!: 3.
The Color blending mode creates an overall toning effect in the image. If you have used gray for the Color Fil l , it becomes another quick way to create a black and-white i nterpretation (Figure 7-96). Figure 7-97. Left: The layer masl(for the Color Fill layer inverted to blacl(. Right: After the masl( has been modified with the Brush tool to blend in the neutral tones and hide the uneven coloring on tile wall.
2 62
I Chapter 7: Color Correction
Figure 7-98. Left: The image before the uneven coloring on the wa!! has been fixed. Right: After applying the neutral color wash to the wa!!.
Figure 7-99. A Violin Concerto in Three Movements: Left: The original image. Center: After applying the Violet Photo Filtel: Right: The final version after locet! Curves adjustments and correcting the color cast all the wa!!.
This technique is obviously not suitable for a l l i mages, but there a re times when using this method of brush-on neutrality works quite well , especially when adj usting an area that might plausibly be a neutral tone, such as this wall (Figure 7-98). And, i f you want to experi ment with having a slight color to the wall, simply double-click the thumbnail for this Color Fil l adj ustment layer and experiment with a new color in the Color Picker. Don't go too fa r to the middle of the Color Picker box or the color you choose wi II become more saturated and less conv incing.
Changing Color Some color correction goes beyond simply correcting an undesi red color cast and i nvolves actually changing color. This can encom pass altering the character of the color, or it can even mean changing a color i n to something entirely d i fferent, or matc h i ng the color between two images. As with other a reas in Photoshop, you can use a variety of commands and techniques to accomplish color modifications.
Altering color with cu rves The ability to target the components of a given color and man i pulate the underlying "color recipes" is what makes Curves so useful. I n addition to correction of unwanted color casts, C u rves is also very usefu l for changing the " flavor" of specific colors. As we saw earlier, the H ue/Satu ration com mand can be used to manipulate i n d ividual colors (for exam pie, shi fting red between i ts com ponent colors of yellow and magenta), but when used with a layer mask, Curves is sometimes more effective for this type of adj ustment. In the i m age of the hull of a n old sailing ship taken in Honolulu harbor, the reddish stripe at the top has a lot of blue in i t, which is pushing the color towards m agenta. Let's explore a C ur ves adj ustment to alter the cha racter of this red stripe (Figure 7- 1 00).
Local Color Correction I
2 63
3.
Switch to the red curve. C md/Ctrl-cli ck the stripe to place a control poi nt on the curve and raise it up to increase the red component. Switc h to the RGB curve. Experiment with l ightening or darkening that area of the i mage. We chose to slightly d a rken the red strip. Click OK. :1
Channell Slue
Il!IZl
86
/
fffil
Channel:! RCB
�-
.. �:
1.
Choose Select�Color Range. I n the Color Range d ialog box, make sure that Sampled Colors is specified at the top of the dialog box. The Fuzziness should be at 40 (the default). Click i n the photograph on the red stripe of the hull to start the selection. Hold down the Shift key (or use the plus eyedropper) and click areas in the i mage that are still black or dark gray in the mask preview in the Color Range d ialog box (Figure 7- 1 0 1 ). Click OK when the mask preview is mostly white over the area of the red stripe. Color Range Sefea:{ "'Sampled Colors :j
Fuuiness:
�
f40I � � � Olnven
Selection Preview: f None
2.
2 64
\Iilith the selection active, add a Curves adjustment layer. The selection will be turned i nto a layer mask. I n the Curves dialog box, select the blue curve. Cmdl Ctrl-click the stripe to place a control point on the blue curve, and then d rag down to subtract blue and add yellow (Figure 7- 1 02) .
I Chapter 7: Color Correction
v
7l .
.
96 �
0
1/1/1,;'loShowCliPPing
/ 0
1/'",'",1 o 'how (II",".
Figure 7-102. The Curves adjustmentsfor modifjling the color ofthe red area on the hull.
As a final step, we chose to use Color Range to create a selection of the water and then adj usted each of the color curves to increase the tropical turquoise character of the water (Figure 7- 1 03 , Figure 7- 1 04).
4.
! ReB
:I
IR.d
V
�,
�
.
( Green
C>
:1
/
.
( Blue
I .
V
.
-
:1
wY
�
:I
Figure 7-101. The Color range dialog box after building a selection maskfor the reddish stripe at the top of the photo.
VI !O3 �
117
:j
{/'
JE! ch7_Harbor_H u l l .j pg
�:
0
I/l/I-,Ioshowcllpping
115
Figure 7-100. The original photograph ofthe hull ofan old ship and the modified version (j) Sf)
{/'
lL
---�
�:
HI
Channel:/ Red
f:1
0
I
-
C>
.
C>
Figure 7-103. Tile Curve adjustments used to change the coloring and brightness ofthe wata
There are two methods for applying the color adjustments: Relative a nd Absolute. A Relative value will add or subtract color proportionally based on the total amount of that color already present. For example, if the amount of yellow in a given area i s 50%, i ncreasing the yellow by 20% will actually add only 1 0% because 20% of 50% is 10%. Using Absolute will result in the actual percentage amount bei ng added, so in the case of the previous example, increasing a 50% yellow by 20% will result in 70%. Relative i s the default setting and i n most cases for making changes t o colors i n RGB images either relative or absolute a re O K to use. The color changes created with the absolute setting a re often much more pronounced than with relative. If you are u sing Selective Color to make subtle color tweaks to CMYK images that will be reproduced on a press, however, then absol ute is the better choice since you have a better idea of what you are getting.
Figure 7-104. Thefinal layers palettefor the Harbor Hull photograph.
Selective color The four sliders for Cyan, Magenta, Yellow, and Black i n the Selective Color dialog box (Layer-New Adjustment Layer-Selective Color) are a h i nt that this command is designed primarily for fine tuning the colors of CMYK images, but it can e ffectively be u sed with RGB fi les. Like the Hue/Saturation command, i t also allows you to choose from a specific color group and then adj ust the component colors. Tn this case, those component colors a re the fou r process inks used for offset pri nting. I n addition t o t h e six additive and subtractive primaries (red, yel low, green, cyan, bl ue, magenta), you can a l so ta rget whites, neutra l s, and blacks, which i s a fa ntastic option when working on RGB files (Figure 7- 1 05).
I n the example i m age of the carnival ride control panel (Figure 7- 1 06), both the red a n d the yellow share component colors. With Selective Color, these can be ad justed separately without the need for a layer mask . While red s and yellows can be a d j u sted i n dividually in H ue/Saturation, you have only the Hue sl ider to make the color adj u stment (we are referring to change s that affect the character of the color, not saturation or brightness). I n the Selective Color dialog box, three d i fferent sliders (cyan , magenta, and yel low) can i n fl uence each color. The black slider will darken or lighten a color and, in this respect, is similar to how the Lightness sl ider affects individual colors in H ue/Saturation. I n the fol lowing example we'll explore how Selective Color can be u sed to alter the character of colors, or change them into entirely d i fferent colors.
Selective Color Options Colors: Cyan:
�
Magenta
� Yellow:
' .
DYellows _Greens DCyans _Blues _Magentas D Whites Neutrals _Blacks
Black: .
0
!t""
p% r% W r--1=1% E:::J%
( ( ( (
OK Cancel Load ... Save...
�Preview
Method: o Relative OAbsolute
) ) ) )
Figure 7-106. The original carnival ride control panel image.
,{)
� ch7_Ca rnivalControlPanel.j pg Figure 7-105. The Selective Color dialog box.
1.
Add a Selective Color adjustment layer to the image. We'll work in broad strokes here just to make it clear how these sliders affect the d ifferent colors i n an image.
Local Color Correction I
2 65
2.
[
Red is the color in itially selected, so we'l l start there. Set the Yellow to -50%. Reducing yellow brings in more of yellow's complement, which is blue, making the red control panel pinke r and more magenta (Figure 7-107).
Colon
,.,.d,
:1 ro-]"
Cvan
---e€r----
Magenta
o
Black:
D
I
I
00
I��--� ��=" I �
00
�
Figure 7-109. With the same settingsfor cyan and yellow, changing the methodfrom Relative to Absolute creates a significant change in the character of the purple colO1:
Figure 7-1 07. Subtracting yellowfrom the reds shifts them more towards magenta, one ofthe component colors in red. 3.
Now decrease the Yellow to -75% and raise the Cyan to +50%, and the control panel becomes d a rker and more purple (Figure 7-108).
4.
Try changing the method from Relative to Absolute to see how that affects the a d j u stment (Figure 7-109). You can a l so experiment with the Black slider to see how it l ightens (negative values) or darkens (positive values) the color. I
C=: '. Reds
5.
Hold down Option or Alt to cha nge the Cancel button to reset. C l ick the Reset button and change the Colors menu to Yellows.
6.
Set the Cyan to + 75% and the Magenta to -75% . Thi s changes the yellow to more of a green ish yellow color (by subtracting magenta, we are adding gree n , a s shown i n Figure 7-110). Changing between Relative a nd Absolute with these settings causes the cyan green color to shift i n to a hyper-saturated fl uorescent green that is so intense it cannot be reproduced in the CMYK colors used in this book. I f you compare the colors on your d i splay to those i n Figure 7-111, you'll see the d i fference.
I
00
� ;::: = '" ''----+---''F=-''I �
Colors:
IDVeltows
Cyan:
Magenta:
Yl!ilow:
Blac.k:
Figure 7-1 08. Subtracting even more yellow and adding in some cyan transfonns the red to a light purple.
2 66
I Chapter 7: Color Correction
Figure 7-110. Adding cyan and subtracting magentafi"Ol11 the yelloUls causes that color to become more ofa greenish yellow.
Colon: IDVellows Cyan:
:1 i7s1.
Magenta:
Yellow:
Black:
Figure 7-111. Using the same Yellow settingsfor cyan and magenta, but changing the method to Absolute creates a bright, fluorescent green that is beyond the gamutfor CMYK.
Selective Color can be useful for subtle tweaks to i mages where there a re d istinct d i fferences between colors, but where you need to increase or decrease a specific color that may be shared by more than one color. I t is especially useful for final color refinements a fter you have converted from RGB to CMYK. The changes in this example have been broad, obvious strokes that a re i n tended to show you how this color correction works. With most images, you will probably not be making such extreme changes. There may be times, however, when you m ight need to a ffect a profound change in a particular color, and becom i ng fam il iar with the different ways you can accomplish this in Photoshop is a worthwhile endeavor.
Figure 7-112. The red-and-white boat. We'll use a variety of techniques to change the red to blue (including the reflections on the water). II!
'1J ch7_RedBoat.j pg 1.
To restrict t h e Replace Color command t o o n l y the a reas we want to a ffect, fi rst d raw a rectangular marquee around the red boat and all of the reflections of the red color that appear on the water. With the marquee selection i n place, choose I mage---> Ad j ustments---> Replace Color.
2.
The controls in the Replace Color d ialog box a re a hy brid of Color R ange a n d H ue/Satu ration. C l ick part of the red trim on the boat, and, when the selection button is enabled, you'll see a selection mask appear in the small preview wi ndow (Figu re 7- 1 1 3). Use the plus eyedropper to add red areas that were not selected with the first click. You can also increase the Fuzziness amount to expand the selected a reas, but we fin d that we have more control leaving i t somewhere between 30 and 40 and j ust using the plus or m inus eyedroppers to shape the selection.
3.
Once you a re happy with your selection mask, move the Hue slider to s h i ft the colors in the ta rgeted areas. If there is a speci fi c color you want to use, click in the Result color swatch to bring up the Color Picker, where you can define a color (Figure 7- 1 14).
Replace Color The Replace Color com mand (Image---> A dj ustments---> Replace Color) is specifically designed to replace one color with a nother, but a very sign i ficant drawback to this feature is that it cannot be applied as an adj ustment layer. As such, we do not fi n d i t very useful. There is a workaround, however, that enables you to combine the useful functional i ty of Replace Color with the Color Range command to create an accurate selection of the areas you want to change. From there, the color can be replaced using an adj ustment layer with a layer mask. Let's take a look at how this can be accomplished (Figu re 7-1 1 2).
Local Color Correction I 267
Replace Color
COlor:
.
00 � � � � Preview
Figure 7-115. Left: Areas on the bacl( ofthe boat were not included in the initial Replace Color selection. Right: Afterfine-tuning the color replacement 117as/(.
e Selection 0 Image Replacement Hue:
n
�
SaturatJon: Ughtness:
n
�
• Result
n
9
FigUl·e 7-113. Thefirst step in using Replace Color is to clicl( a color in the image and begin to build a selection mas/(for that COlOI: Replace Color
Color:
•
00 � � � � Preview
e Selection 0 Image Replacement --------,
58 1 ______ _______I-_'_
� � ' H"'
�
Saturation: Ughtness:
�
c
n
•
Result
Figure 7-116. Left: The gunwales on the right side ofthe boat showing some holes il1 the selection 171as/(. Right: After adding those areas to the replacement color mask. 5.
Click the Save button in Replace Color and save the settings (Figure 7- 1 1 7). Remembe r where you save them because we'll access these i n the next step. I f you want t o use the same color when w e create a n adj ustment layer, make a note o f what your Hue setting is. For the blue that we chose, the Hue value was - 1 58. After the settings have been saved, click the Cancel button to exit Replace Color wi thout making any changes. I f you had clicked OK, then the color change would be a pplied to the i mage with no way of adj usting it late r.
o
Figure 7-114. Move the Hue slider to choose a replacement COI01: To access a specific COI01; click in the Result color swatch to open the Color Pid(e,:
The really useful aspect of Replace Color is that by actually changing the color, you can see where the selection needs to be i mproved. When we first changed the recl to blue, for example, there were some obvious areas at the back of the boat and a long the gunwales that had not been i ncluded in the i nitial selection mask (Figure 7- 1 1 5 and Figure 7- 1 1 6). 4.
268
Now that the color change is i n place, use the plus eyedropper (or Shift-click) to click areas where the color has not been replaced. Note that some areas of the recl reflections on the water a re also being included. This is good, but these areas will probably have to be fi ne-tuned a l i ttle later.
I Chapter 7: Color Correction
o �tCllon a Im�gt Rtpl�ctmtnl ------�
� n �---------pw "n II ro-! Ruull --.... s _ . ---
UghlntSl;
c
Figure 7-117. Thefinal selection mask created in the Replace Color dialog box is saved so that it can be loaded into the Color Range dialog box. The blue reflections are obviously too pUiple, but we'll fix those in a later step.
Using Replace Color Settings with Color Range 6.
Now choose Select--->Color Range. Click the Load button and fi n d the Replace Color file (the file extension is .axt). The selection mask that you created in Replace Color is loaded i nto the Color Range dialog box. Click OK to create the selection (Figure 7- 1 1 S).
7.
Ad d a Hue/Saturation adj ustment layer. Move the Hue slider to the same setting that you had i n Replace Color, or to whatever color looks good, and click OK (Figure 7- 1 1 9).
:I � [iSl � � � a:1IlJ
_ _ _ _ ''-- Inverse to change the selection so that everything except the boat is selected. Press the Delete key to fi l l the selection with black. Choose Select-> Deselect. You now have two Hue/Saturation layers that affect the boat and the reflections separately (Figure 7- 1 22). Figure 7-123. Top: The version beforefine-tuning the boat's reflection. Bottom: A more believable reflection.
Match Color The Match Color command ( Image->Adjustments->Match Color) is used to sample and m atch color c haracte ri stics between d i fferent i mages, or w it h i n selected a reas of i m ages. Additionally, sampled color statistics can be saved as a reference file and then applied to other i m ages. It is very effective for matching color between two i m ages of the same subject where the coloring may be d ifferent due to a change in l ighting cond ition. I t can a l so be used with selections to create a match between elements that are d i fferent colors. We'll explore both scen a rios in the following examples.
Figw'e 7-122. The edited layer mask! that apply color changes separately for the boat and its reflection on the wate1:
2 70
I Chapter 7: Color Correction
2.
Choose I mage-> Adjustments->Match Color. The fi rst thing you need to do is go to the lower part of the d ialog box and specify the source (Figure 7- 1 2 5). (We think this is a bit backwards; the fi rst control you need to access should be at the top of the d ialog box). Open the Source menu and choose the mailbox- l photo. Match Color will do i ts magic and, for these images, it works pretty well (Figure 7- 1 2 6).
vai lable as an adj u stment layer r, so to preserve yo u r options and ructively, it's best to either wo rk o n a
f i l e o r on a d u p l i cate layer.
ch Color G lobaJly. The two close-up photographs of Match Color
he old mailbox detail i n Figure 7- 1 24 have d ramatically d i fferent color balances because the afternoon sun illum inated one side of the mailbox and the other side was in the shade. Match Color can fix this pretty quickly. The first step is to decide which i mage will be the source that you will use as a color reference. For this exam pIe, we'll use the warmer sunlit i m age as the source.
o Neutralize Image Statistics -. Use Selection In Source to Calculate Colors _
Use Selection In Target
r ch7_Mailbox D... :j ) layer: ! Background :� ( Load Statistics . ) ( Save Statistics...
10
Calculate Adjustment
Source:
Figure 7-125. The Match Color dialog box.
Figure 7-124. Two close-ups ofan old mailbox-one side in the JUn, the other in the s/Jade. !0 \1)
..!EJ ch7 Mail boxDeta i l 1 j pg _
-
.
..!EJ ch7 Mail boxDetail 2 j pg _
1.
-
.
Arrange both photos in Photoshop so you can see them side by side with View->Screen Mode->Standard Screen Mode. On the bluer, cooler mai lbox i mage (the one that shows the h i nge), make a copy of the Background layer by choosing Laye r->New->Layer via Copy, or use the very handy Cmd-JICtrl-J shortcut.
Figure 7-126. The coolel; shady mailbox photo (bottom) matched to the sunny-side mailbox photo (top).
Loca l Color Correction I
271
Match Color Image Options. The I mage Options let you fi ne-tune the color match if such tweaking is needed. Lu m i n a nce i s a brightness control, and Color I n tensity will boost or red uce the color saturation of the origi nal image. Fade acts l i ke an opacity control, as if the color match effect were on a separate layer, a llowing you to fade the effect and c reate a blend of the matched color and the origi nal appea rance. Neutralize attem pts to apply a neutral color balance to the i mage, though that will, of course, cancel out any color matching. Neutral ize can be useful, however, for removing color casts on some photos (Figure 7- 1 27) and can be very e ffective in neutralizing u nderwater shots.
Using Save Statistics. Save Statistic, color recipe of the i mage so you can appl) without actually having the original source ., You don't need to be actively matc h i ng two image, statistics. I f you have a fi le that you want to use as a refere . . fi l e , just open i t up, access t h e Match Color dialog box, a n d save t h e statistics. I n practice, the SLlccess rate using saved statistics can be hit or miss, and the result is often q u i te d i fferent from what you were expecting. It genera l l y works bes t with images that are similar i n content, overall brightness, and contrast. For exam ple, the statistics from a close-up shot of a corroded beach shower produce i n teresti ng results when applied to the mailbox detail photos (also beat-up pieces of old metal), as seen i n Figure 7- 1 28, but the resu lt is quite d i fferent when applied to a shot of the USS Bowfi n at Pearl Harbor (Figure 7- 1 29). The Fade option in Match Color is often useful i n these situations for sca l i ng back t h e i ntensity of t h e effect. In Figure 7- 1 30, a Fade value of 50 created more of a strong warm i ng filter effect.
Open the teapot and the vase i mages. To cut to the chase, we have a l ready prepared an alpha channel for the vase. I n the vase i mage, choose Layer-'>Dupl icate Layer. In the Channels palette for the vase, Cmd/Ctrl click the vase al pha channe l to load i t as a selection (Figure 7- 1 32 ). If you want to create the selection you rsel f, the Quick Selection tool works very well on the vase.
2.
In the teapot i mage, make a Lasso selection of the side of the teapot (Figure 7- 1 33) . I t does not have to be accurate to the edges of the item.
Figure 7-130. By using a Fade setting of 50, the effect wasfine-tuned to something resembling a wmmingfilter. (, " ) Match C o l o r with selections. For those times when you
need to affect only a portion of an i mage, such as when matching color between clothi ng or products, selections can be used with Match Color. As with most procedures that involve selections, the success is often determi ned by how good the initial selections a re. I n Match Color, the source selection does not always have to be super precise, but you do have to have an accurate selection for the ta rget i m age. To explore this technique, as well as some of the problems that can arise, we'll apply the coloring of a small teapot to a ceramic vase (Figure 7- 1 3 1 ).
Local Color Correction I
273
XOlllno,.s..le"k>New Adj ustment Layer-->Curves. I n t h e N e w Layer d i alog box, c l i c k t h e check box for Use Previous Layer to C reate Cli pping Mask. This will apply the curves adj ustment only to the layer i m mediately below it.
2.
I n the Curves dialog box, lower the center section of the RGB curve to darken the vase a bit (you should have the teapot i mage open and visible so you can refer to it as you make the adjustment). Place a poin t on the upper part of the curve that represents the h ighlights so that the bright highlight on the shoulder of the vase is not darkened. (This is the main reason why we chose not to a d j ust the brightness using the Luminance slider i n Match Color; that highlight would have been da rkened as well.)
3.
Open the channel selector and choose B l ue. Place a point on the center part of the blue c u rve and lower it slightly to add a bit of yel low to the vase. Click OK to apply the adj ustment (Figure 7- 1 42 and Figure 7- 1 43).
( RGB
j:j
( Blue
",.----
rV V 6
.
�:I
---
j
•
jLavers x I Channels I Paths I �Normal I :J opaclty:[Iooill lock: fOIJ]±OO FiII:IIOO"H �
,
�
�
6
� - x
Curves 1
I� [t].[]� [t] �
I
The color correction techniques we have explored so fa r in this chapter have all been centered on RGB files. The reason most photogra phers deal with R G B files is that their digital cameras produce them and it is the best format to send to photo-qual i ty i n k j et printers. The other color mode that is often useful in some situations is Lab (pronounced as three separate letters, L-A-B, and not like a place where scientific experiments a re conducted). The structure of Lab color i s fundamentally d i fferent from the other color modes because i t separates brightness or l u m i nance from the color v a l ues in a n image. In both CMYK and RGB, the color channels contain not only color i n formation but also brightness values. The separation of brightness from color m akes some types of color correction much easier in Lab since you don't have to worry that a correction to the brightness w i l l change a colo r val ue, or that a tweak to one of the color channels will alter the brightness of that color. But while this separation of l u m inance from color can make certain types of corrections easier, the familiar methods that work in R G B ra rely yield the same results in Lab, and sometimes they can create ghastly changes that prom pt one to quickly reach for the Cancel button. An u nderstanding of the basic Lab terrain is necessary before you go exploring. I n the final section of this chapter, we'll tinker under the hood of the Lab color space to see how to apply Curves to a Lab i m age, and then we'll take a look at two d i fferent scenarios where working in Lab can produce results that wou l d be m u c h harder to achieve in RGB.
III
"',kg,ounO
.... ft. ill f!>. O
The Power of Lab
@
�
Figure 7-142. Thefinal curve to fine-tune the results ofMatch Color: The circled cu'ea on the RGB curve shows the anchor point placed in the highlights so t1ley are not dar/cened along with the rest of the vase.
F(gure 7-143. Left: The resultsfrom using Match Color and oval selections ill both the JOUI"Ce and tile tCllget. Right: The version afterfine-tul1ing with a Curves adjustment.
The Power of Lab I
277
We are s p e n d i n g o n ly a s m a l l part of two chapters to cover the tip of the p rove rbial Lab i ceberg. An e n t i re book could be written about this topic, and, i n d eed, o n e a l ready has been. If after read i n g th is y o u a r e i n trigued b y c o l o r corrections i n Lab, then we reco m m e n d yo u take a j o u rney with Dan Marg u l i s and h i s defin itive explo ration of wo rking i n Lab, Photoshop Lab Color: The Canyon Conundrum
& Other Adventures in the Most Powerful Colorspace
(Peachpit Press,
2006). To d e lve i nto creative color
e n h ance m e nt with Lab p l ease see 'Exploring I m age Mode B l e n d i ng' in Chapter 8, "C reative Colo r".
Lab Basics In RGB i m ages the brightness of the gray val ues in the channels d i rectly correlates to how m uch of that color is present in a given part of the i mage. Lighter tones of gray indicate more of that channel's color, and darker values ind icate less of that color (Figure 7- 1 44). The brightness values of all th ree color channels combine to create the overall brightness and contrast in the i mage. In an image in Lab mode, the brightness and color values are completely separate. The former is controlled in the L channel (L is for lightness or l u m i nance). Because the L channel contains only brightness information, i t resembles a normal black-and-white version of the image. While the L channel looks pretty fam il ia r based on what we're used to seeing in an RGB file, the color channels in a Lab file are d i fferent beasts entirely. Since they contain only color information and do not carry any brightness detail, they can appear very strange when you first start working with them. The a channel controls the magenta and green color balance, and the b channel controls the yellow and blue balance. J ust as with an RGB i m age (Figure 7-1 45), the l ightness of the gray val ues in the a or b channel tells you something about the color that it represents, but i t uses a completely d i fferent code. An exact 50% gray in a Lab color channel indicates no color at all. So if there is a n area in both the a and the b channel that is 50% gray (a value of 0 in Lab-speak), then that is a neutral tone. In the a channel, gray values l ighter than 50% represent a magenta color and tones darker than 50% gray indicate a green color. In the b channel, l ight gray values represent a yellow color while those tones darker than 50% gray represent a blue tone. Positive val ues i n dicate warmer colors (magenta, red, yellow) while negative values represent cooler colors (green, cyan, blue). In both channels, the darker or l ighter a tone becomes, the more i ntense and saturated the color w i l l be.
278
I Chapter 7: Color Correction
Figure 7-144. An RGB image and the three color channels that contain the brightness and color information. (u Sf)
Figure 7-145. In Lab mode, the brightness /lalues are contained in the L channel (L /01' lightness), and the color in/olmation is totally separate in the a (magenta-green) and b (yellow-blue) channels.
Working with Curves in lab If you a re new to working in Lab mode, then making adj ustments in the Curves dialog box is a mixture of the com fortingly fam i liar and the totally foreign . The familia r part is making adj ustments t o t h e L channel. Since i t contains only brightness and contrast i n formation, it behaves very similar to the RGB curve in an RGB file (Figure 7- 146). The foreign part comes when you have to venture into the a or the b channel to ad j ust the color balance of a n i mage (Figure 7- 1 47). The type of curves that you m ight use in a n RGB file will rarely produce good results in Lab mode. Before we get to the main tutorial i mages, let's explore some cu rves basics for working in Lab.
FIgure 7-147. The basic opposing color relations/IlI)S ofthe a and b curves (the color shading on the channel selector menu was addedfor th is iflustration).
The importance of the center point An i m porta nt thing to k now about working with the a and b channels is the meaning of the center point on the cu rve grid. As long as the c u rve passes through this center point, there will be no cha nge to the color bal ance of the channel. To accomplish such a move you need to move the corner points horizontall y along the top or bottom by the same amount. Doing so will a lways ensure that the curve runs through the center poi nt. The effect that this has on your i mage is to increase the i ntensity or saturation of colors; the steeper the cu rve, the more i n tense the colors (Figure 7- 1 48 and Figure 7- 1 49).
(.
1 :1
-----
ro�=m==========
Figure 7-148. The importance ofthe center point: By moving the corner points the same distance along the top or bottom of the curve grid, the curve will always pass over the center point ofthe grid. This wilf increase the intensity ofcolors but will not change the overalf color balance.
FIgure 7-146. The L channel contains only bnghtness and contrast information. The curve adjustments you ma/(e to the L channel are similar to what you mIght do to the RGB curve.
The Power of Lab I
2 79
:.
:1
Figure 7-149. Left: The original image. Right: After applying the a and b curves shown at thefar right. The colors have been inte12Jified, but the overall color balance has not changed. Specifically, the 50% gray m'ea at the bottom ofthe image has not changed in value. Values of 0fOl' the a and b channel indicate a neutral tone.
Controlling brightness and contrast with the L channel The best way to become familiar with how making adj ustments in Lab color can help with tricky i mages is to give it a try. I n the fol lowing tutorial the problem photo has brightness, contrast, and color cast issues. The scene was photographed in the late afternoon shadow of the Andes in Peru, which yielded a low-contrast i mage. Adding to the overall blue tint from the shaded l ight is the fact that farmers were burning their fields nearby and clouds of bluish smoke were d ri fting through the scene. Some targeted adj ustments in Lab color mode work very well for this i mage (Figure
7- 1 50).
JEJ ch7_Lab-Plowing.jpg 1.
Open the file and convert i t to Lab by choosing I mage->Mode->Lab Color.
2.
Add a Curves adj ustment layer. I n the Lightness curve, move the shadow corner poin t horizontally along the bottom of the grid to a poin t where the histogram detail begi ns (approximately the third vertical grid line in from the left). Do the same thing to the highlight poi nt in the upper-right corner: Move it along the top of the grid to where the histogram shape begins (one vertica l gridline in from the right). The image i n Figure 7- 1 5 1 is looking better al read y !
Figure 7-150. Plowing thefield: A low-contrast original with a bluish cast caused by the late afternoon shadows ofthe Peruvian Andes and smo/(e from nearby burning fields. On the right, the Lab-corrected version. .If)
280
I Chapter 7: Color Correction
Now, if you're thinking that you could accomplish the same thing with the RGB curve, you would be correct. . . sort of. I n order to adjust the brightness and contrast i n an RGB file, the color channels have to be adjusted. In the worst cases, this can sometimes lead to color shifts. I n the case of this image, no new color shifts have been introduced , but the existing blue colorcast is more appa rent and the overall result is not as good as the same adj ustment applied to the Lightness channel (Figure 7-1 52).
Lab Figure 7-151. Moving in the endpoints ofthe L channel improves the contrast ofthe image and even helps remove a bit ofthe blue cast.
C>
C>
Figure 7-153. Moving the a and b channel corner endpoints in by equal amounts increases the color saturation of the image but also makes the blue cast more noticeable.
Removing a color cast with the a and b channels
Figure 7-152. A similar adjustment applied to an RCB ven'ion ofthe file is certainly an improvement over the original, but the bluish color cast is even more pronounced. 3.
From the channel selector at the top of the Cu rves dialog box, choose the a channel. Move both corner endpoints in one-and-a-half grid l ines. Switch to the b channel and make the same adj ustment. This helps boost the color saturation. You can see it most in the hide of the brown bull and in the green fol iage at the bottom of Figure 7- 1 53.
4.
The i m age still looks a bit too blue. Since we're already in the b channel, which is where blue can be i n fl uenced, we'll warm up the i mage a bit. Move the highlight corner poi nt in a bit more to the second vertical grid l ine. This subtracts blue by adding a bit of yellow (Figu re 7- 1 54).
5.
N o w for some fi n a l fine-tuning. I n t h e i m age, Cmd/ Ctrl-click the trees above the stone wall to place a control poin t there on the b curve. Now add several other control poin ts on either sicle of this point. These will serve as lockdown poi n ts so other a reas of the curve clo not move. Retu rn to the poin t that represents the trees above the wall. Use the arrow keys to nudge i t to the left ancl up a bit. This adds more yellow to the m i d tones. Switch to the a chan nel and clo the same thing to acid m agenta-red to the mid tones. The move i n the a channel will be much more subtle and slight than the one i n the b channel (Figure 7- 1 55).
The Power of Lab I
281
Dodging the b u ll's face
Figure 7-1 54. Moving the highlight b channel endpoint in to the second vertical grid line subtracts some ofthe blue.
6.
As a final step for this i mage, we l ightened the face of the black bull. Add a new C u rves adjustment layer. Click the bul1's face to see where that point is located on the curve. D rag that poi nt up and l ighten that area to taste. You m ight also want to place a point down near the corner and d rag that down to preserve the contrast. The enti re image will be a ffected , but we'l1 fix that in the next step. Click OK.
7.
Choose I m age�Adj ustments-->I nvert to switch the layer mask from white to black. Choose the Brush tool and an appropriate-sized soft-edged brush tip (we used a I SO-pixel brush). Press D to set the default colors which wil1 place white in the foreground swatch. I n the Options bar for the B rush tool, set the opacity to 30%, and begin painting over the areas of the bul1's face that you want to l ighten (Figure 7- 1 56).
Figure 7-155. Two subtle adjustments to the midpoint ofthe a and b channels add yellow and magenta to the midtones to wcmn up the nnage.
Figure 7-156. A second CW've layer with a layer masl( opens up the shadows all the black bull's face by raising the lower part ofthe Lightne,'" curve. 2 82
I Chapter 7: Col or Correction
Build a Custom Color Correction Filter The fact that the color values are separate from the brightness and contrast let you do some i n teresting thi ngs in Lab that are not possible in RGB. One technique based on this property of Lab is to add a Photo Filter adjustment layer, sample the color that is responsible for the cast, and then use i nverted values for the a and b chan nel to correct the cast (Figure 7- 1 57).
2.
Add a Photo Filter adjustment layer. Click i n the color swatch to open the Color Picker. In the i mage, click the area where you placed the color sampler point to change the Color Picker to display this color. r n the numeric fields that show the Lab values, invert the n umbers for the a and b channels. Do this simply by adding a minus symbol i f the number is positive (as is the case with this image), or by removing a m inus symbol i f it is a negative number (Figure 7- 1 59). Click OK.
3.
Back i n the Photo Filter dialog box, raise the Density setting unti l the color cast begins to be neutralized. For the typewrite r i mage, we ended up using a density of 80%. Click OK (Figure 7- 1 60).
J!
L:
Co J! M : Y: K' a-bit
a:
b:
8-bit +.
X: Y:
NI L: J! a: b:
Figure 7-157. A strong color cast is corrected by using a custom Photo Filter layer ba>'ed on inverted Lab values. \/J
40 14 34
10
W: H:
Figure 7-158. A color sampler is placed on an area that should be neutral or close to neutral. Inverted versions oftile a and b values will be used to create a custom color correction filter.
"EJ ch7 Typewrite rLAB.j p g _
1.
Open an i mage with an obvious color cast, or you can use the typew riter i m age featured here. Choose the Eyedropper tool and Shift-click a n a rea that should be a neutral tone to place a color sampler poin t there. The top of the typewrite r j ust above the manufacturer's name is a good place in this i m age (Figure 7- 1 58). Jot down the values for the a and b channel.
The Power of Lab I
283
Photo Filter
Use
I f y o u l ike t h e e ffect of a n adj ustment layer, b u t would l i k e i t t o b e j ust a l i ttle bit more, duplicate t h e l a y e r and see what a double dose looks l ike. I n the case of the typewriter image, we used the Cmd/Ctrl-J shortcut to duplicate the Photo Filter adj ustment layer. Then we double-clicked the layer thumbnail to reopen the Photo Filter dialog box and changed the Density setting for the duplicate layer to 60% (Figure 7-1 6 1 ).
( OK ) ( Cancel ) � Preview
O Col
[2Sl%
Density:
Q
� Preserve Luminosity
Select filter color:
til •
a s:
Cancel
)
( Add To Swatches )
0 H �'
a s:
OK
( (
1fs"J% [57]%
o R: lsoJ OC: [loOl O s: [144l .f lE6490 I
Color Libraries
O L: f39l O a: Fl4l Ob: i=34l C: 196 1%
fS8l % v: fUl % K: i4l%
M:
Figttt'e 7-159. After clicking the color swatch in the Photo Filter dialog box, the specified color in the image is sampled, and then the a and b values are edited in the Color Pickel: In this case, 14 and 34 were changed to -14 and -34.
Figure 7-161. This shows the result ofduplicating thefir>·t Photo Filter layer and lowering its Density setting to 60%.
When to Use lab I n the hands of a skilled technician or artist who is familiar with the ins and outs of worki ng i n Lab, that color space can be used with many types of i m ages: the good, the bad , and especially the ugly. But for those who are new to Lab, i t is probably best suited for i m ages that have significant "issues" you are having trouble with using more conventional solutions. Lab is great at cutting through hazy indistinct shots, creating custom color correction " filters," and, when used correctly, working wonders on old scans that suffer from low contrast and an abysm a l color cast.
Warming Fiher (85)
o Colo r:
Density:
•
.,
Cancel
� Preview �" �
� Preserve Luminosity
Figure 7-160. After configuring the inverted color valuesfor the a and b channel, increase the Density setting as necessary.
2 84
I Chapter 7: Color Correction
The primary strength of Lab is the separation of brightness and contrast from the color v al ues. And while the Lightness channel is easy enough to figure out, worki ng with curves in the a and b channel can take a l i ttle getting used to. Though it m ight seem like rocket science at times, it's really not. I t's j ust color. And with a l i ttle practice, even the m ysteries of Lab color can eventually seem as commonplace as RGB.
A Journey
Through the landsca p e of Color
Color is all a round you. You see it every day. I ndeed, you have been looking at i t your entire l i fe. But looking at color and seeing color are two very d i fferent thi ngs. To look is to cast your gaze around you and to notice. To see is to go beyond looking and come closer to comprehending and understanding the color. And this comprehension of color, how it affects the delicate balance in a scene, and how you can best capture i t with a camera and work with it i n the digital dark room, is of critical i m portance to a visual artist. It is the d i fference between what is and what is possible.
We feel truly fortunate to be able to use some of the most wonderfu l d igital i m aging software in the world in our digital dark rooms. But a t the heart of our journey through the landscape of color, the n u m bers and math and nuances of software fade away, and we a re left with the basic observations and questions that we started with when we fi rst looked at a photograph we had taken. And these observations and questions are not fra med in software termi nology. They are the broad strokes of simple inquiry i nto what the i m age is now and how we m ight make it better. What is and what is possible.
Figure 7-/61. Some photographs l'equire very little adjtwment and recognizing when not to edit is aim an imp01·tant digital darkroom skiff. Thi,8-second exposure taken in Acadia National Park has only received minor contmst and j'aturation adjustments. Va riations t o v iew a color ring around their images. The problem with Variations is that i t works only o n 8-bit images and is n o t a n image adj ustment layer feature. This means you either have to d uplicate the background layer to work non-destructively or use i t as a quick preview to apply similar effects with other Photoshop features such as Photo Filte r or Hue/Saturation adj ustment layers. A more flexible and non-destructive color preview can be achieved by adding a ful l spectrum gradient to your image, as explained here.
1.
Choose Edi t-->Preset Manager, and use the pull-down menu to select Grad ients. Use the fly-out menu to select Spectrums. After releasing the mouse, make sure to click Append to add the Spectrums gradients to your grad ient palette without replacing the original gradients.
•
•
The 80 and L B B are cooling filters. Blue in color, you can use them to correct p ictures with strong yellow to orange colorcasts created by taking the picture in tungsten or candlelight. The L B B blue has the slightest h i n t of m agenta in i t and i s based on the Fuji LDD filter, while the 80 is based on the Kodak filter. The 82 filter is a milder cooling filter. Use it for waterfalls or snow scenes to turn them slightly blue, emphasizing the coolness of the subject.
Photos hop Photo F ilters o ffer a subtle power to i m prove i mages. Since the P hoto Filters are adjustment layers, you have the additional advantage o f being able to adj ust their strength with the layer opacity and layer blending modes, as well as control where the corrections take place with layer masks. We're not suggesting that you take bad pictu res o n purpose, but now you will be able to improve those photos that didn't turn out the way you expected.
2.
Select Layer-->New Fill Layer-->Gradient, and then select the Spectrum gradient, as h ighl ighted i n Figure 8-6. We find that Linear or Angled works best for this exerCise.
3.
Click O K and change the layer blend mode to Color and the layer opacity to 30% to preview the image with a wide variety of color tints, as seen in Figure 8-7.
4.
I n this example the blue/violet tinted areas gives the i mage a quiet strength. To apply this specific color to the entire i m age, i n the layer palette, click the Background layer, and use the Eyedropper tool set to 3 by 3 pixels to sample the desired tint.
5.
Discard the Gradient fill layer and add a Photo Filte r adj ustment l ayer. C l ic k t h e color swatch i n t h e Photo Filter i n terface, and then use the Eyedropper to click on the foreground color i n the tool bar.
6.
Adj ust the density. In this example we used 35% to create the final i mage, as seen in Figure 8-8.
Using the G radient F i l l layer to preview color tints is a quick, non-destructive techn ique to explore color tinting.
Working with Hue I
291
Foreground to Background Foreground to Transparent Black, White Red, Green Violet, Orange Blue, Red, Yellow Blue, Yellow, Blue Orange, Yellow, Orange Violet, Green, Orange Yellow, Violet, Orange, Blue Copper Chrome Spectrum Transparent Rainbow Transparent Stripes
Figure 8-6. Use the Spectrum gradient to preview color choices.
Figure 8-7. Viewing the image with the low opacity Gradient enables you to preview how colors influence the image.
292
I Chapter 8 : Creative Color
Figure 8-8. Slightly cooling the image imbuej' the bild with a subtle undertone a/nobility. \!J KI,
Warming Images Whenever there a re people in an i mage, adding a subtle warming effect with the SI Photo Filte r will warm the skin tones and make the i mage more appealing, as seen in the before and a fter example in Figure S-9.
1.
I n t h e Layers palette, c l i c k t h e "create n e w fill o r a d j ustment layer" icon, and choose Photo Filter.
2.
Choose Warm i ng (SI ) and a d j ust the density. In most cases when working with skin tones we keep the warming fi l ter setting between 5-20% density.
JEJ ch08_2 g i rls.jp g When u s i n g the Ph oto Fi lter keep the Preserve Lu m i nosity c h eckbox ch ecked to i n s u re that the i m age d e n s ity will not c h a n ge as t i nts are a p p l i e d .
Adding warmth selectively Many times you'll want to warm a specific section of the i m age. Start with a selection, and Photoshop transfers the selection to the layer mask when you add the adj ustment layer. Only the active a reas (which appear white in the layer mask) will be a ffected, as shown in Figure S- 1 0 in which the door was warmed to contrast with the coolness of the winter day.
Figure 8-9. Warming the image livens up the sl(in tones. Hn �'l'IIJl!i!l
\h"';;
Figure 8-10. Just slightly warming the door and increasing its contrast givej· it a j·tronger visual presence. 'J /(I'
Worki ng with Hue I
293
1.
To select clearly defined colors and shapes such as the door, use the Quick Selection tool and d rag i t around the entire door in a circular pattern to create the active selection.
2.
I n the Layers palette click the "create new fill or adjustment layer" icon and choose Photo Filter. Use Warming (85) in this example and increase the density to 50%.
3.
To increase the effect of an adj ustment layer we often cycle th rough the blending modes to what effect they will have. In this example, selecting Soft Light with a layer opacity of 50% pumped up the door to enhance the wood beautifully as shown in Figure 8-1 1 .
To learn how to blend colors based on tona l values to create more dramatic effects, see the upcoming "Sophisticated Color Blend i ng" section.
Cooling Images Cooling an image imbues the scene with serenity and strength. I nd ustrial, u rban, a n d snow scenes are the classic candidates for cool ing effects. I n the example seen i n Figure 8- 1 2 adding a Photo Filter with Cooling (80) at 40% density cooled the image n icely. To increase the e ffect, we changed the blending mode to Overlay to create the final i mage.
Figure 8-12. The cooler version with the Overlay blending mode enhances the mood ofthe winter day. B lending Options ancl, i n this case, the desi red blue color is cooling the darker a reas of the man and the hawk of the i mage too much. B y moving the This Layer shadow slider far to the right, the blue effect is concealed (Figure 8-2 1 ). Option!Alt and d rag the shadow slicler to split the slider and create an i mperceptible transition.
The key to worki n g with Blend I f i s to u nderstand that mov i n g t h e active T h i s Laye r s l i d e rs conceals the effect, w h i l e m ov i n g the U n derlying Layer s l iders reveals t h e layer's tones or colors t h rough the T h i s Laye r.
Worki ng with Hue I
297
be_sky
Figure 8-22. Enhancing the image with color and contrast creates an intriguing 1non2ing 111.ood. l[ "l/lk Nrd(c/n}{]11 lleoo.cIln, Opllon� Cunom
UDroCIShadow Ow.cr�_ !Jo.lItfClow UlnnnDIM
A�udll ;fl [1001"
� � (NewStyle ... ) �Prevtew
1l LJTururt
8T'�IW.PltYSh_1� OU'f'!.u.ntHidnlnem
J ..... (jColoto..tt..,
QVtaOf!.tUkll;duUrKU
UCud",,,,o-by Ohtl.."Ovorby
JEJ ch08_Chysler.jpg 1.
Photographed agai nst the l ight on a warm summer morning, the air is thick with h u m id ity, which reduces the i m age contrast. To push the l ess i m portant buildings on the left out of v isual i n terest, add a Curves adj ustment layer and, as seen i n Figure 8-23, push the shadow detail down to d arken the buildi ngs.
�lzft-r.
Und..
Figw'e 8-21. Adjusting the Blend lfsliders allows the blue to be enhanced, whichframes the man and the bird beautifully.
Neutral Density and Col or Bl ends Enhancing photogra phic color is emotional ly similar to painting, in which a painte r applies a brushstroke and then reflects, reacts, and responds to the brushstroke by add i ng compl imenta ry brush strokes. I n the process the i mage grows and changes and gai n s an identity of its own. [n the case of sophisticated color blending, worki ng with layers, masks, and blend modes enables you to experiment with the image and, admittedly, the fi nal outcome often creates itself. It is best to start off subtly and build up the effect gradually. I n the example seen in Figure 8-22 , Mark Beckelman started with an early morning shot of the C h rysler Building in Manhattan and, with a few layers, created the m uch richer final image.
298
I Chapter 8: Creative Color
I:I�
� � � (o.nlon.... ) e!!pm1.....
Figure 8-23. Dm'kening the shadows makes the building visually less important.
2.
To enhance the morning l ight, load the pastel gradients by selecting Edit-Preset Manager, use the pull-down menu to select Pastels. Click Append to add them to the G radient palette.
As is often the case, the creative process requires that you work flexibly and, in this i nstance, Mark wanted to add density to the sky and warm the buildings to better reflect the new color of the l ight.
3.
Add a G radient F i l l layer and use the Linear Yellow, Pink, Purple pastel gradient, which, as you can see in Figure 8-24, obliterates the enti re image. Since this gradient is on its own l ayer, the Multiply blend mode "vith 23% opacity will enhance the image perfectly (Figure 8-25).
1.
.....
For�v'OUndIOh(kgrOund
I
�
To warm the build i ngs while not a ffecting the rest of the image, we need a selection tha t excl udes the sky and i ncluded the buildi ngs. Of course, there a re many ways to make a selection, but in instances that a re so clearly defined, you can let Photoshop do the work. Cmd/Ctri-click the icon of the green channel in the channel palette to activate the tonal values of the channel into an active selection, as seen in Figure 8-26.
Forl!910undto Trlnsp.lrent
Blut....ellow.Blue
•
Orlngt,Yellow,O'lnge Vloltt,C;r«n,Of"oIng"
� YI! IIOW.Vtol"t.Ofilngt.BIUI!' I
�
coPpl!r
IIJJJ
J
�
Spt!ClfU m
"'tllow.Pink,G'Un
...
Yellow, Pink, Purp�
Figure 8-24. The pastelgradients mimic softer light.
Figure 8-26. Using the channel tonality to effectively create (/n accurate selection. 2.
In the Layers palette, click the layer beneath the grad ient, and add a Photo Filter adjustment layer with the standard orange. As you can see in Figure 8-27, the l ighter portions of the image are now orange, which is exactly the opposite of the desi red e ffect.
Figure 8-25. Changing the layer blend mode to Multiply and adjusting the layer opacity creates an early morning mood.
Working with Hue I
299
Figure 8-27. The channel luminosity is the start ofa very accurate layer mask. 3.
Input Levels:
I[
Choose I mage�Adj ustments �Invert to i nvert the tones of the l ayer mask, and then increase the mask contrast with Levels, as seen in F igure 8-28. This makes the buildi ngs warmer without changing the sky. When you try this tec h n ique on your own i mages you may want to use a Gaussia n B l u r fi l ter of 2-5 to soften the transitions of the mask. I n this example, it's not needed, but we often use it.
[9Ol
Output Levels:
•
[OJ
"
[Q.5Ol
1l
Cancel
) ) Auto ) [228] ( Options... ) [Zl)'Il] Smart Object->New Smart Object via Copy. Although this seems l i ke the long way to d upl icate a l ayer, s imply d ragging a Smart Object to the new layer icon creates a duplicate that references the original Smart Object. Making a change to t h is d ragged copy i n Camera Raw w il l also apply the change to the original. You can a l so use the Layers palette context-sensitive menu by Cmd/right-cl icking the name of the Smart Object, and selecting New Smart Object v ia Copy.
5.
Double-clicking the layer icon w i l l open Adobe Camera Raw, a l lowi ng you to adj ust or change the white balance o f the d u pl icated file.
6.
As the layer palette in F igure 8-35 shows, Katrin exper imen ted w ith each white balance and named the layers after the W hi te Balance sett ing she used.
Creative Color Temperature I
305
I n terestingly enough, while creating the different S mart Objects, she thought that the tungsten setting was the best, but a fter some consideration, the fluorescent setti ng adds the appropriate lonely mood to the i mage of the tombstone. T h is goes to show that you have to see the mod i fied image before j u mp ing to concl usions.
Figure 8-36. Photographed 011 a November afternoo11 and reorchestrated with creative color to better render the original perceptions. fopSh� UIn"tfSNdow UOu". G\ow U
I
UConto...
O �ndlnl..!Or ur.( U I l GrOUp
I!f
UTi....,,.
� a'-"d CI'llP'dL.lVf" u C.rO"'p U»lln U ColoI ov.,l&v U GtMl""l o..rtay
I!!
....·.. (Jow
U ltwllnd [mbaU
Tranlp�lfIl('jSh�I LaVtr
U�lln
uytrMa.k HIdIl UftReveal All to increase the canvas size to contain the l a rger i mage.
1 2. To offset the softness of the long exposure, click the topmost layer and press Cmd-Option-Shift-E on a Mac, or, Ctri-Alt-Shift-E on a PC, to me rge all visible layers up onto a new layer. Select Filter-'>Other---> H igh Pass with a radius of 5. Change the layer blend mode to Soft Light to tighten the image edges to insinuate sha rpness and create the final image (Figu re 8-57).
1 1 . After adjusting the i mage resolution, add a new layer and name it "retouch." Use the Clone Stamp and Heali ng B rush to repa i r the sky and to tone down the extreme h ighl ights that were caused by the strong spotlights a long the river edge, as seen in Figure 8-56.
Figure 8-5 7. The High Pass layer set to Soft Light adds needed crispness.
The abi l i ty to revisit a n i mage with new ideas and u pdated software is one of the prima ry reasons that we both shoot in the raw format. As our creative vision grows we can rework images to better render our perceptions, and as the raw acqui re software is i mproved we can reprocess files with better resul ts. Figure 8-56. Retouching the artifilcts and multiple light streaksfocuses the image.
Creative Color Temperature I
31 3
Explori ng I mage Mode B l en d i ng Most color i mages that photographers work with a re i n RGB mode, but this doesn't m e a n we're l i mited t o having only three channels to work with. Dan Margulis has expounded on this idea i n his books and semi nars, and he mainta i ns that every color image can be composed of ten channels: RGB, CMYK, and Lab. Combining R G B and Lab i mages with i mage mode blending tech niques can make a lackl uster i mage pop off the page. Best of all, this technique requires only a l ittle experimentation to create breathtak i ng results very quickly. We find it works especially well when enhancing greens, golds, and l ight modul ation i n l andscape photogra ph y.
1.
D u p l i cate Image Duplicate: foutain before.psd As: Ifountai n_LAB _
Dupl icate Merged Layers Only
( I (
OK
)
Cancel
)
Figure 8-59. Adding "LAB" after the image name alleviates confusion in the process.
2.
Wi th the d u pi icate file acti ve, select I mage-> Mode-> Lab Color.
3.
Return to the original file and select I m age->Apply I mage. Use the Source pull -down menu to choose the dupl icated Lab color file, and i n the Channel menu select the "a" channel. Set the Blending pull-down menu to Softlight to add a subtle tweak of contrast and color.
4.
Experiment by changing the channel to "b" and checking I nvert, as seen i n Figure 8-60.
Blending Color and Light This technique requires you to dupl icate the file and convert the dupl icate to Lab, and then blend the "a" or "b" chan nels back onto the origi nal RGB image with Apply I m age. To extend the potential, you can take advantage of the H istory B rush to further refine the i mage, as explai ned here and seen in Figure 8-58. Each step provides a lot of visual feedback for you to respond to, making this technique very easy to experiment with while producing creatively satisfying and d ramatic e ffects.
Select I mage->Duplicate, and we suggest adding the letters "LAB" after the file name (Figure 8-59) to help reduce confusion when using Apply I m age.
-1EJ ch08_fo u ntain.jpg
Figw'e 8-60. Experimenting with the "a " or "b " channel and the Invert checkbox often produces excellent results. 5.
Figure 8-58. Lab blending brings out the texture in the wall and enhances the colors in the wate;: /,, 1 314
I Chapter 8: Creative Color
To increase the d ramatic e ffect, change the Blending to Vivid Light, as seen in Figure 8-6 1 . In many cases you' l l need to reduce the Opacity when using Vivid Light. But i t is worth it to experi ment with each "a" or "b" channel, checking the I nvert box, and seeing what the Overlay, Soft Light, H ard Light, and Vivid Light create.
r Sourct' : fountaln.LAB
"::,",--::::-:- �",;-_-j I
...
.-
°t
illl ;:; !
Loy", ' ,k,,,,,,". : ::.nnt!1: c""b___== � -=:J � l nYt! . L
� f!j Prtvitw
Figure 8-61. Vivid Light increases the contrast and punch of the colol: 6.
After clicking OK use Cmd/Ctrl-Z repeatedly to see the before and after versions. I f there are a reas in the "before" version that you like more, use the H istory Brush to paint them back i n . Click the col umn next to the desi red H istory State, and paint from the past into the present, as we d i d in this example to bring back the details in the water (Figure 8-62 ).
Figure 8-63. Lab blending brightened the color,·, adding vibrancy to the image. A.
1.
Select I mage-Dupl icate a n d a d d t h e letters L A B a fter the duplicated fil e name.
2.
Retu rn to the original file and dupl icate the background layer to al low for greater flexibility.
3.
Select I mage-Apply I m age. Use the Source pull down menu to choose the duplicated Lab file and the Channel menu to select the "b" channel. Use the Blending pull-down menu to select Overlay to en hance the contrast and color, as seen in Figure 8-64.
Figure 8-62. Use tile Hi.ftory Brush to restore the detail il1 the watel:
Color Bl ending with Lay er Mas ks Combining va riations of the same color-blended i mage with layers and layer masks enables you to refine specific parts of an image, as seen in Figure 8-63, in which Katrin first applied an overa l l blend to en rich the color and l ight, and then added an additional blend to deepen the color of the water.
'1l ch08_sail boats.jpg
Figure 8-64. The initial blend enhances the color and the blue wate!:
Exp lori ng Image Mode B lending I
315
4.
Duplicate the blended l ayer, select I m age-'>Apply I m age, and use the same settings with the "b" channel and Overlay. In this example, too much contrast has been added to the background (Figure 8-65) .
T h e beauty of the L a b b l e n d i n g is the resu l ts a re often surprising and inspiri ng. I n a l l honesty, we often can not pred ict what will happen based on which channel or blend mode we choose. After years of massaging every pixel i n an image, this serendipity is sincerely welcome a n d c reatively refresh i ng.
Create Cross-processing Effects When working with color fi l m , we both enjoyed cross processing color fil m , which entailed p rocessing slide fil m i n color negative chemistry o r color negative fil m (shot with a 2 stop over-exposure) i n £6 chemistry. This resul ted i n higher contrast, v e r y saturated, a n d intriguing fal se color, as seen in Figure 8-67. Using Photoshop to c reate a similar effect, as seen i n Figure 8-68, requires a lot less l uck and can be refined at any time. The th ree primary characteristics to emulate a re : contrast, color shifts, and h ighlights and shadows that lack detai l .
"CJ ch08_CrossProcess.jpg Figure 8-65. Duplicating the layer and repeating the blend strengthens the effect nicely. 5.
Add a layer mask by clicking the layer mask button on the bottom of the Layers palette, and use the gradient tool set to Black, White to d raw the gradient from j ust below the horizon straight up to the top of the image, as seen in Figure 8-66, to block the added contrasts and still let the water pop.
Figure 8-66. The gradient on the layer mas/( conceals the unwanted contrast of the image background, focusing the viewer's eye on the blue water and the saitboatJ". Figure 8-67. Two examples oftraditionally cross-processedfilm.
31 6
I Chapter 8: Creative Color
u
When experimenting with cross-process i n g effects, w e reco m m e n d wor k i n g w i t h colorful and graph i cal i mages with large s u rfaces. In o u r expe rience, i mages w i t h fi n e detai l t e n d t o be less s u ccessful for this tech n iq u e becau se the added contrast obl iterates the d e l i cate detai ls.
Figure 8-69. Start by increa,.ing overall image contrast.
2.
C ross-processing requi res ruining the real color, which is ach ieved by changing each color channel. Add a new Cu rves adj ustment l ayer used to mod i fy the i ndi vidual color channels. I n the red c hannel increase the contrast by p u l l i ng the d a rk e r 3/4 tones down and the l ighter 114 tones up, as seen in Figure 8-70. :I l!d �
"'UR !CuSIOIfI
r-;r / ' I l � I � "'"
A/-#
Qulpuc
//�
. Input:
Figure 8-68. Cro,.,.-proce,.,.ed in Photo,.hop.
1.
"
07
Q s.- CflPIIlng
-/ '
Start by adding a C u rves a d j u stment layer and, i n Photoshop CS3, use the Preset Strong Contrast. I f you're working with a n earlier version o f Photoshop you can achieve a similar effect by adding a strong S curve that emulates the curve seen in Figure 8-69. Figure 8- 70. fncrea,.ing the red d1Clnnel contraJ·t.
Create Cross-processi ng Effects I
31 7
3.
[ n the green channel pull the mid tones up to offset the strong red cast (Figure 8-7 1 ).
5.
: (c"nom OIil1n'J. � -
Add a Solid Color layer and choose a deep rich yell ow. In this example, Sean used RGB 255, 2 1 5, 1 00 and, a fter clicking OK, c hanged the Blend Mode to Color and reduced the opacity to 20%. To create the e ffect seen in Figure 8-73 , Sean used the B l end I f slider to allow the deeper values of the underlying l ayer to show th rough. Of course you can experiment with the color and the opacity, as it feels right to you. OK
� (NtwStv\e ... ) 6!,,,vltw
I
Figul·e 8-71. Lightening the green channel to createfalse colol: 4.
Use the blue channel to make the shadows and h ighligh ts m urky and give them a color cast by moving the shadow poi nt up and the highlight down. Bumping the midtones, as seen i n Figure 8-72 , changes the color of the light towards the bl ues.
om ---====m l·crnlI .,.: nrc;;;ti: � .. P
.7J
[!;J
� � """'"
I�=======:===:�::;:������} f��------�� D� ( �
O Only Web Colors
. lffd9.. I
(Add To Swatches ) (",,' ) ( Colorllbrarin ) Oc f90l O " rI 0.0 [67l c fill · Mo flSl_ ,, !73l. " fill ·
Figure 8-73. The modified yetlow wlorfill layer creates the classic cross-processing effect. 6.
Figure 8-72. Flattening the blue shadow and highlight values.
318
I Chapter
8:
Creative Color
Add an additional Curve layer and pull down the red curve to push the i mage i n to an otherworld l y color mood. [ n this example, Sean also pulled down the blue channel ever so sl ightly, to let the yellow shine th rough, as shown i n Figure 8-74.
7.
To deepen the shadovvs further, add a new Cu rves adj ustment layer, use the Preset Strong Contrast, and reduce the l ayer opacity to 75% (Figure 8-75).
Cross-processing may not be everyone's cup of tea, but for pop-art, graphical-effect a ficionados it can be very appeali ng and, w hen used carefully, can i mbue images with an edgy, retro feeling.
It's
Colorful Worl d
a
o magic recipe, secret sauce, "make good" command, or perfect picture setting exists that would work on every image. Enhancing color is a very subjective endeavor that involves some trial and e rror to learn what works best for your i mages. Start with the basic warming and cooling techniques, and then move on to working with blend modes and layer masks to make your i mages sing out in ful l color revel ry. Paul Gauguin said i t best, "It is the eye of ignorance that assigns a fixed and unchangeable color to every object beware of this stumbling block. "
l
..�,
[dlrL cc"-o"":
8 C...... DI'�1'0III�' ----
Figure 8-74. Aclj'usting the red and blue color challllelJ pushes the image into the cyan and yellow spectrum. Curves
Prntl: !Cunom
�R::; G'====-!I==
===:
Ch�n tl: !
IlL]
� �
(Options... )
g Previtw
Output'
Input:
,t'
o Show CHpplng
G CUIW OlsPlaVOptlons -----
Figure 8-75. Adding an additional overall contrast Curve ties the image together by blocleing the shadows and forcing the highlights to a lacle of detail.
It's a Colorfu l World I
319
C reative E n h a n ce m e n ts As photographers, we make i mages so that we can bring them back to our d igital darkrooms and revi s it that fraction of a second captured in a photograph through the filter of creative consideration. The enhancement of the images we make begins with the photographic act itse l f, t hrough the use of filters, tripods, and specialized exposu re tech n iques, or waiti ng at a location for a few extra hours until the l ight gets better. Creative enhancement cont inues in the d igital d a rk room even if that enhancement is subtle and m inor, designed only to create a n image that i s as fai t h fu l as possible to what we remember we saw (or what we think we remember we saw) at the moment we pressed the shutter button. Even though the pixel is steadi l y replacing the silver halide crystal as the stardust with which we make our images, we stil l look to the world of fil m, dark rooms, and texture for inspiration to e n hance our p hotographs. In this chapter, we explore d igital techniques that address: •
Optical and Film Effects
•
Darkroom Special Effects
•
Adding Texture
•
The Distressed I mage
•
Creative Edge Treatments
In a way, the focus of this book has been centered on this one topic : how to enhance your images-and your experience with the i mages-in the d igital darkroom. We h ave covered a lot of ground and have delved deep i n to many fundamental and essential tec hniques for getting the most out of your photographs and m a k ing them look better. In t h is chapter we treat the photograph as a story, and consider how we m ight embell i s h and enhance t h a t story t o transform t h e photograph i nto something j u s t beyond t h e l i teral document of w h a t was in front of t h e lens w h e n t h e s hutter opened.
Chapter 9
Optical a n d Film Effects Over the years, photographers have used fi ltration techniques and film choices to create i mages with a specific look. Many photographers tell us that digital captures look "too perfect" and smooth, and they look for ways in the digital darkroom to take some of that polish off so the i mages look more film-like. In this section, we take a look at digital techniques for creating classic photographic visual treatments such as diffusion and soft focus, film grain, and the black-and-white infrared image effects.
Diffusion and Soft Focus In the past we used glass filters on the camera to create a d i ffusion or soft focus effect. Now, we use Photoshop. Although such filters a re still available for the camera lens, in most cases we would rather capture a sharp image and decide later ifit needs a little soften i ng. Adding d i ffusion or soft focus with Photoshop is easy, but removing it when it is part of the original image is not. Photoshop provides a variety of ways to create soft focus and d i ffusion effects, some involving a filter effect and others that use blurring and blending modes.
3.
Now, set the blending mode to Lighten and leave the Opacity at 60% . This c reates a more subtle soft focus effect that is similar to using a soft focus filter on the camera lens (Figure 9-2).
4.
For experimental exploration, c hange the mode to Overlay and leave the Opacity at 60%. As is always the case with Overlay, there is a n increase i n contrast and color saturation. In this i mage, the natural colors c reate a warming effect (Figu re 9-3). If the increase i n contrast is too pronounced, set the mode t o Soft Light.
5.
Click Cancel and then Delete the Smart Filte r layer (j ust the filter e ffect and not the Smart Object layer of the farmhouse).
Gaussian Blur -!EJ ch09_DiffusionFarm.jpg 1.
Choose Filter--Convert for Smart Filte rs t o turn the active layer into a Smart Object, and then choose Filter--Blur--Gaussian B l u r. In the Gaussian B l u r dialog box, set t h e Radius t o 8, and t h e n click O K .
2.
In the Layer panel, double-click the blending options button to the right of the filter name to bring up the Blending Options dialog. Set the Opacity to 60%. This c reates a blend of the blurred layer with the original sharp image (Figure 9- 1 ). Vary the opacity as needed for the particular photo.
Figure 9-2. Setting the blending modefor the Gaussian Blur layer to Lighten alld the opacity to 60% refines the diffusion effect colltrast.
� � � Prtvitw
� � Prt\lj"w
Figure 9-1. The original image is on the left. The diffusion effect on the right was created by adding a slight Gaussian BIU!; and then lowering the Smart Filter opacity to 60%. , Sf) 322
I Chapter 9: Creative Enhancements
1. When worki n g with Photoshop CS2 or earl ier, p lease d u p l icate the backgro u n d layer to experiment with filter effects.
� � � Preview
T h e m a i n layer i n t h e fi le should sti l l b e a Smart Object. [ f not, choose Filter->Convert for Smart Fi lte rs. Choose Filter->Blu r->Su rface Blur. Set the rad i us to 20 and the Th reshold to 75. This will c reate a very strong blur that w i l l obl iterate some of the fi ne details in the i mage (Figure 9-5). We'll fi ne-tune the effect in the next step. Cl ick OK.
The " s u rface" i n S u rface B l u r refers to areas i n the i mage that do n ot conta i n edge details. At lower settings than we a re u s i n g here, this allows you to b l u r areas that a re tonally very s i m i lar, or that have grad ual transitions, leavi n g hard, d i sti nct edges u n affected.
� � � Preview
Figure 9-3. Setting the blending modefor the Caussian Blur layer to Over/ay and the opacity to 75% creates a diffusion effect with more contlrlst.
Remember that Smart Filter l ayers come with a layer mask that can be edited to fu rther refine where in the image the filter effect shows up. We'l l explore that in the next example.
Surface Blur T h e Su rface Blur fil ter c a n be used for subrle soften ing or for a more pronounced d i ffusion effect (Figure 9-4). In this exa mple, we'l l take the latter approac h .
Figure 9-5. The initial effect created by the Stllface Blurfilter is quite strong.
Figure 9-4. The 5Uifoce Blur diffusion technique adds a painterly effect.
Optical and F i l m Effects I
323
2.
Double-click the blend i ng options button to the right of the fi l te r name. Lower the Opacity to 80% to soften the blur. Click OK.
3.
Click the layer m ask for the Smart Filter l ayer. Choose the B rush tool, set black as your foreground color, and lower the brush opacity i n the Options Bar to 30%. Brush over areas where you want to fu rther minimize the soft focus e ffect created by the Surface Blur filter. I n the i mage of the farmhouse, we masked out some of the bl ur on the house and the tall tree (Figure 9-6).
1.
I f you're sti l l using the same image from the previous example, throw away any Smart Filter effects. If you a re using a different image, be sure that the layer has been converted for Smart Filters using that option in the Filter menu. Choose Filter--Distort--Diffuse Glow.
2.
The G rain iness slider controls how much grain is apparent. The Glow Amount i ncreases the glow around the l ighter a reas i n the image. The Clear setting removes the glow from d a rker a reas i n the image. Set the G raininess to 6 and the Glow and Clear Amounts to 8, and then click OK. In addition to the bleached-out d i ffusion effect, there is also a muti ng o f t h e colors i n t h e photo (Figure 9-8).
1;1_ """"""' • Ou, """"""
..
.�
Figure 9-8. The settingsfor the Diffuse Glowfiltel: 3.
The farmhouse has become a bit too light. We can fix that by painting on the layer mask for the Smart Filter. Click the layer mask to make it active. Choose the B rush tool and make black the foreground color. Set the Brush opacity to 30%. Paint on the mask in the areas where you want to minimize the D i ffuse Glovv effect.
4.
Now let's add another filter e ffect. Click back on the Smart Object layer to make it active. C hoose Filte r-> Blu r--Gaussian Blur. Set the radius to 1 2, and then click OK.
5.
In the Layers palette, double-click the blending options button to the right of the Gaussian Blur fi l te r name. Set t h e blending mode t o Overlay a n d the Opacity to 75%, and then click O K (Figure 9-9).
Figure 9-6. The SUlface Blurfilter layer is set to 80% opacity and a layer mask is used to prevent the house and parts ofthe treefrom becoming too soft.
Diffuse Glow Diffuse Glow is actually two filters i n one si nce i t a llows you to add grain as well as to create a bright glowing effect on l ighter a reas in the image. I t can be used as a stand alone fil ter e ffect or i n conjunction with blending modes to customize the e ffect (Figure 9-7).
Figure 9-7. Combining the Gaussian Blur and Diffuse Glow filters, blending modes and a layer mask created this effect.
324
I Chapter 9 : Creative Enhancements
JEJ ch09_sepiaesque.jpg 1.
Add a H ue/Saturation adj ustment layer. Turn on the Colorize checkbox i n the lower-right corner. Set the H ue to 38 and the Saturation to 1 5 (See Figure 9- 1 1 ). Click OK.
2.
Set the blending mode for the H ue/Saturation layer to Overlay, and lower the Opacity to 70% (See Figure 9- 1 2) . l f the contrast is increased too much by the Overlay blending mode, you can either lower the opacity more or try the Soft Light blending mode.
Figure 9-9. Thefinal effectfrom combining the Diffuse Glow filter with the Gaussian Blurfilter set to Overlay and 75%.
Sepiaesque Diffusion Toning The tech nique described h e re is one that Sean developed for use with color photographs to create a faded, sepia flavored color palette combined with a soft d i ffuse glow (see Figure 9- 1 0). (The name sepiaesque is his own description of the technique and does not refer to any specific Photoshop feature.) Like many tech niques for modi fy i ng the character of color or contrast in a n image, the magic of blend i ng modes is a key i ngredient. This techniq ue is not a one-size-fits-all procedure, however, and images that a l ready have a lot of strong contrast may not fare as well as more low-contrast photos. Since everythi ng is created using separate layers and adjustment layers, the l ayer masks can be modified easily to customize the effect for each i n d ividual image.
Figure 9-10. Sepiaesque diffusion toning. Before and ajteJ: '
Figw'e 9-11. Thefirst Hue/Saturation layer creates an overall sepia tone.
Sf)
Optical and Fi l m Effects I
325
Figure 9-12. The effect ofthe sepia Hue/Saturation layer after setting the blending mode to Overlay and lowering the opacity to 70%. 3.
Use Cmd/Ctrl-J to d u p licate the H ue/Saturation layer. A double dose of this l ayer creates an interesting e ffect, but on some i mages i t is l i kely to produce too m uch contrast and shadow details will become too dark, as can be seen in the blue cover over the beach chai r. Lower the opacity to 30% and set the blending mode to Normal. The main purpose of this layer is to fu rther tone down the color saturation and also to add some additional sepia seasoning (F igure 9- 1 3) .
Figure 9-1 3. The second Hue/Saturation sepia layer is set to NormaL at 30% opacity.
326
I Chapter 9 : Creative Enhancements
4.
Make a merged copy of the existing layers by pressi ng the Option/Alt key while selecting Merge Visible from the Layers palette submenu. Keep the Option/Alt key down until you see the new layer thumbnail a ppea r in the Layers palette.
5.
With the merged duplicate layer active, choose Fil ter-> Convert for Smart Fil ters to turn the layer i n to a Smart Object.
6.
Choose Filter->Blur->Gaussian Blur and set the radius to 1 0. Cl ick O K (Figure 9- 1 4) .
7.
Cha nge the blending mode for the blu rred layer to Overlay, and lower the opacity to 50% or 60%. I f the e ffect seems too contrasty or the shadows get too dark, try using Soft Light for the blending mode.
8.
For this i mage of the beach chair, the blue cover has gotten a bit too dark. To mod i fy it, we'll use a layer mask. With Layer I selected (the merged, blurred layer), click the Add Layer Mask button at the bottom of the Layers palette. Choose the B rush tool from the Tool palette, and use a l arge, soft-edged brush to paint with black at 40% brush opacity over the blue canopy. The fi rst H ue/Satu ration layer is also causing some of this da rk ness, so click its layer mask and paint over the canopy again to further restore some of the original color and tone (Figure 9- 1 5).
Si nce the d i ffusion blur is applied as a Smart Filter, you can also use its layer mask i f there are a reas i n the i mage w here you want to scal e back the d i ffusion.
Figure 9-14. AppLying the Gaussian BlurfiLter to the 111elged copy {ayel:
Figure 9-15. Thefinal image, after modifyillg two ofthe layer masl(S to lightell the blue canopy.
Adding Film Grain Si nce the invention of fil m , its manufacturers worked to refine their fil m formulas to reduce gra in, a visible a rtifact of the granular structure of the light-sensitive silver halide crystal s that a re used in film emul sions. But over the years, many photographers developed an appreciation for certain types of grain and would speci fically use certain types of fil m or special processing techn iques to add grain to their images. Digital cameras don't produce gra i n, but it can be added in Photoshop (Figu re 9- 1 6). As with a l l such e ffects in the d igital darkroom, we try to c reate a fi l m grain e ffect in a non-destructive way.
Figure 9- 16. The original is on the top, and the version with a film grain effect on the bottom. \/)
The actual noise a m o u nt you decide on w i l l d e p e n d on t h e size o f the i m age you're wor k i n g on and h ow m uch grain y o u want. Val u es over 1 0 t e n d t o g e t very g r a i n y a n d begin t o obscu re i mage detai ls .
'1J ch09_filmgrai n.j p g
layers )(
� �
The Add Noise filter 1.
2.
� Preview
Choose Fi l ter--->Convert for Smart Filte rs, and then choose Filter --->Noise--->Add Noise. [n the Add Noise dialog, set the Distribution to Gaussian and the amount to between 5 and 1 0. Choose the Monochromatic option (Figure 9- 1 7). The gra in effect can now be modified by a l tering the opacity of the Smart Filte r effect or by painti ng i n the layer mask to tone down the appea rance in certain a reas.
I Normal
Channels
I
;:i
Paths
�OO%J3 F))): �OO% J3
: Op"lty:
Lock: DY�
Iii __
.� ......
-D
.
,o
-,
Sman 'iJ'e"
", I
• Add NOIse 100%
13 Amount:
�
[lcfI"
8
.... "'.LD �.
1iI
�
I
Olstrlbu,lon
U
o Uniform o Gaussian
IS1 Monochromatic
Figure 9-17. Adding 1I0ise non-destructively using a Smart Filter layel:
Optical and F i l m Effects
I
327
Creating softer grain to mimic an enlarged negative
Black and White I nfrared
While the Add Noise fi l ter does a good job of simulating a grainy look, the effect i s a bit too sharp for some people. The solution to this is to put the effect on a separate l ayer so that a slight b l u r can be i ntroduced to soften the grain. The result more closely resembles the appearance of actual fil m grain i n an enlarged print, w h i c h is softened a s t h e negative is enla rged.
In addition to the visible spectrum of l ight, i n frared fil m s or digital sensors used w i t h a n i n frared fi lte r a r e also sensitive to the i n frared portion of the spectr u m . This renders scenes quite d i fferently from the way we a re used to seeing them. Vegetation reflects a great deal of i n frared l ight that the human eye cannot see, and green fol iage is recorded in ethereal white or very l ight tones. Blue skies and water a re rendered as very dark tones since they reflect l i ttle or no i n frared l ight. There is something magical about photographing the actual i n frared l ight i n a scene and never knowing exactly how the i mage might come out that lends an appealing air of mystery and serendi pity to the process.
1.
Option/Alt-click t h e N e w Layer button at the bottom of the Layers palette. I n the New Layer d ialog, set the Mode to Overlay, and then click the check box for "Fil l with overlay-neutral color (50% gray)." Click O K a n d zoom in to 1 00% (View--->Actual Pixels).
2.
Choose Filter--->Convert for Smart Filters to turn the gray l ayer i n to a Smart Object. From the Filter menu, choose Noise--->Add Noise, and then configu re the dialog settings as needed to produce the desi red amount of grain. C l ick OK. The Overlay blend mode, which does not show areas of 50% gray, now shows only the noise in the image. If the noise effect seems too pronounced with Overlay, try using the Soft Ligh t blend i ng mode, l owering the layer opaci ty, or mod i fying the l ayer m ask.
3.
To add a slight soften i ng to the noise e ffect, choose Filte r---> B l ur--->Gaussian Bl ur, and set the radius to 0.4 pixels. Click OK (Figure 9- 1 8).
The blu rring effect is very subtle, and you have to look closely to see it. Also, the radius val ue you use will vary depend i ng on the size of your image. Larger files can take a s lightly higher radius. I f you've ever fel t that the Add Noise filte r produced "grain" that was j ust a bit too sharp, then this is one way to take some of that edge off.
Although Photoshop does not have a one-stop "infra red fi l ter," it does have a preset in the Black & W hite adj ustment dialog box that can get you started, and there are add itional techniques you can use to transform a color photograph i n to a pretty convincing i n frared image. The process starts with knowing the visual qualities of a n i n frared image and then finding the best ways to d u pl icate those in the digital darkroom. Figure 9- 1 9 shows a n i mage that was photographed on black-and-wh i te i n frared fil m . The enlarged detail section reveals the noticeable grain that is common in i n frared fil m . Figure 9-20 is a photo taken with a d igital SLR that has been specially modi fied to shoot only infrared images. I n both photos, the fol iage is rendered in classic, l ight silvery tones with a bit of a glow. The d igital infrared shot, however, has no grain to speak of.
Figure 9-18. On the left. the noise is applied normally. On tile right. a slight Gaussian Blur has been applied to the noi.,e layer to soften the grain.
3 28
I Chapter 9: Creative Enhancements
JEJ ch09_BWinfrared.jpg
Black-and-white adjustment for the initial infrared conversion
Figure 9-19. This photo ofa eucalyptus grove was photographed all 35111111 blad(-and-white infraredfilm. ,/)
1.
The fi rst step is to convert the color photo to black and white. Open the Add Adj ustment Layer menu at the bottom of the Layers palette, and then choose Black & Wh ite.
2.
r n the B lack & W hite dialog box, open the Preset menu at the top and choose I n fra red. This gets the settings headed i n the right d i rection, which is why we're using it, but it al ways needs some customization. For example, i n the i mage of the eucalyptus grove, the i n frared preset renders the h ighl ights in the green grass much too bright (Figure 9-2 2).
Figure 9-20. The same location photographed with a digital SLR modified to shoot only blad(-and-wI1ite infrared.
The steps to create a black-and-white i n frared interpretation of a color image, as in Figure 9-2 1 , require a black-and white conversion that renders the green colors very l ight, some glow a round the highlight a reas, the addition of some grain, and a final contrast adj ustment. Let's get started.
•
F6il"
•
r:rl"
n·
Figure 9-22. The infraredpreset make>' the grass mud1 too bright.
Figure 9-2 1. The eucalyptus grove photographed in color and converted to black-and-white injiared in Photoshop.
1/'
Optical and Fi l m Effects I
329
3.
4.
5.
For this photo, we stil l want the green grass to have the classic silvery white look, but we also want to see a bit of detail there, so we lowered the Yellows and G reens slider to + 1 50 each . We also increased the Cyans setting to + 2 50 as this hel ped to l ighten up areas i n the background where some fol i age i s visible. I n the I n frared preset, the Reds a re at a negative value, which makes no sense at all, since red a reas, such as the reddish bark on the tree, will record as very l ight i n an actual i n frared shot (refer to the fil m and d igital i n frared exposures i n Figures 9- 1 9 and 9-2 0). We raised the red value to + 1 50, which helps to brighten the ribbon of reddish bark that i s peeling off the tree (Figure 9-23). Fi nally, we lowered the Bl ues to -75, si nce blue val ues typically do not register very bright i n i n frared. Click OK to apply the black-and-white conversion.
The actual B lack & White adj u stment settings you use w i l l vary d e p e n d i n g on the natu re of the i mage you're wor k i n g with and also you r own preferences for how you want the i nfrared conversion to appear. Use the I nfrared Preset to get started, but c u stom ize the s l i d e rs to get the look that you want, rather than j u st relying on a p reset.
CY.anI:
•
Add the glowing highlights The next step i s to create a glowing e ffect a round the highl ight areas. As with so many Photoshop tech n iques, there i s more than one way you m ight go about this. We' l l show you two d i fferent ways. Both create t h e requ isite glowi ng foliage, but each has i ts own flavor.
Method
1:
The Diffuse G low filter
1.
Make t h e B l a c k and W hite adj ustment l a y e r active and hold down Option/Alt while selecti ng Merge Visible from the Layers palette submenu. Keep the Option/Alt key down until you see the new layer thumbnai l appear in the Layers palette. This w i l l c reate a n e w layer at t h e top of t h e stack t h a t i s the merged result of all the visible layers.
2.
Choose Filter......Convert for Smart Fil ters. Then choose Filte r ...... D istort...... Di ffuse Glow. Set the G raini ness to 5 (we'll be adding a separate grain layer later i n the process), the G low Amount to 1 0, and the Clear Amount to 1 5, and then click OK.
3.
The initial effect is m uch too bright (Figure 9-24). I n t h e Layers palette, double-cl ick t h e blend i ng options button to the right of the fil ter name. I n the Blending Options dialog box for the D i ffuse Glow fil ter, set the Mode to Overlay and the Opacity to 50% (Figure 9-25). The use of the Overlay blend mode here accom plishes two thi ngs: i n boosting the contrast i t adds t o t h e glow effect, but i t also da rkens t h e shadow a reas of the scene, which is typical of an i n frared exposure. Click OK.
[»0]" .
- f.Tsl" W&lI'ntu.
•
FiOOj"
Figure 9-23. This adjustment tones down the brightness in the grass and lightens the red values (note the peeling barl( on the tree), which would be rendered light by an actual infrared expoJUre.
33 0
I Chapter 9: Creative Enhancements
Figure 9-24. The initial resultfrom the Diffuse GlowJilter is interesting but much too bright.
Figure 9-26. Creating a diffuse glow effect by blurring only the green channel on the copy laYe!: Figure 9-25. Modifying the Diffuse Clawfilter effect.
Add a grain layer Method
2:
Blurring the green channel. V.ie learned this
method for creating an infrared glow from Martin Evening. It uses a duplicate layer of the original color image and appl ies a subtle blur to only the green channel. Here's how it works: 1.
After applying t h e i nitial Black & W h ite adjustment as described earlier, c reate a copy of the background layer (Cmd/Ctrl-J). I n the Channels palette, click the thumbnail of the green channel to make it acti ve.
2.
Choose Filter-B l u r-Gaussia n Blur, and apply a 5-pixel blur to the green channel. I f you want more obvious flaring i n the i mage, then use a h igher setting. Click OK (Figure 9-26).
3.
Choose Edit-Fade Gaussian Blur, and set the blending mode to Screen and the Opacity to 25%. Click OK.
Photographs made with black -and-white i n frared fil m a re generally quite grainy, and that is one of the classic qualities associated with the black-and-wh ite i n frared i mage. To keep the grain separate from the other visual components of the i mage, use the Add Noise filter on a 50% G ray layer set to Overlay, as described in the previous section on "Adding Film G ra i n . " Zoom i n to 1 00% (View-Actual Pixels) for the most accu rate eval uation of how much noise to add. I f you want the option t o b e able t o change the Noise settings, then apply it as Smart Fil ter. Modi fy the grain effect using the layer opacity slider or painting i n the l ayer mask i f you need to m i n i mize the grain i n certain a reas.
Customize the contrast with curves To fi ne-tune the i n frared effect, you al most a l ways need to a pply some contrast c u rves, either globa l l y or locally. For the photo of the eucalyptus grove, we added a curves adj ustment layer to the top of the layer stack and slightly increased the contrast with an S-c u rve (Figure 9-2 7).
Optical and Fi l m Effects I
331
mistakes are merely a n annoyance and a tem porary setback that can be remedied by making a new print or m ixing a new batch of chemicals. Sometimes, however, mistakes result in an unexpectedly altered i mage that can look very interesting. For some photographers, the accident they fi rst curse becomes a creative catalyst that starts a new visual Journey.
Sol arization
Figure 9-27. Thefinal image after adding grain and a curves layer to increase con trast.
Photoshop infrared vs. real infrared As this example shows, with the right tech niques i n Photoshop, and t h e k nowledge of w hat real black-and white i n frared looks l i ke, you can create a pretty con vincing effect. One area where the fau x infrared does not match the real exposures, however, is i n the hanging leaves in the background a reas. In the real i n frared exposu res-both fi l m a n d d igital-these leaves show u p as a m uch l ighter tone, since they were reflecting i n frared l ight that the fil m or filtered camera sensor recorded. In the version made from the color i m age, these background leaves did not have m uch green i n them, so they were not lightened by the initial Black and W hite adj ustment.
Darkroom Special Effects In a traditional darkroom , the i mage is c reated through a mixture of l ight, chemistry, and time. The amount of l ight, the specific d il u tion and temperature of the chemical sol utions, and how long the film or pri nt is immersed in those solutions all contribute to the fi nal appearance of the image. Some photographic e ffects have their roots i n the special alchemy of the chemical dark room. While creati ng these effects traditionally was often a process of trial and error, creative experi mentation is much easier in the digital darkroom . In this section we'll take a look at the digital versions of some classic darkroom effects.
Creative Accidents Anyone w ho has ever worked i n a traditional darkroom knows that m istakes a re sometimes a part of the darkroom experience. If the chemicals are m ixed i ncorrectly, or the l ight is tu rned on at the wrong moment, prints-or worse, unexposed fi l m-can be ruined. Most of the time, these
332
I Chapter 9: Creative Enhancements
Solarization is the common name for a process that is technically known as the Saba tier effect, after Armand Sabatier, the French photographer and scientist who is generally cred ited with its discovery in the 1 860s. Exposing the fil m or paper to l ight d u ri ng the chemical development process creates the Saba tier effect. The result is an i mage that appears to be half negative and half positive. The mid tones and shadows appear normal, but the highlights are reversed and turn dark, leaving a thi n, white outline around the original highlight edge. As with many such "discoveries," a mistake most l i kely led to the first solarized print and, since accidentally turning on the l ights is one of the most common darkroom m istakes, we imagine that many photographers in the ea rly years of the medium "discovered" this i nteresting visual treatment. In the 1 930s, the noted surrealist photographer Man Ray elevated solarization to a creative technique that he used in many of his most famous photographs.
The Solarize filter Photoshop has had a Solarize fi lter for many years (Filter-> Stylize->Solarize), but it's a simple formula effect with no options for customizing how i t works.
JEJ ch09_solarize.j pg 1.
To apply the filter non-destructively, choose Filter-> Conve rt for Smart Filters. Then choose Filter-> Stylize->Sola rize.
To get insight into j ust what the solarize effect does, take a look at the Histogram palette after you apply the filter (Figure 9-28). All the values in the image a re below the midtone point. The filter takes any val ues at level 1 28 or above and inverts them, remapping them to below level 1 28. Original i mage values below the midtone point are not changed. Curves is the key to this e ffect, both i n terms of enhancing the basic Solarize fil ter as well as creating your own c ustom sola rize treatment, which we'll get to shortly. If you do choose to use the can ned fil te r effect, you can a l ways i m p rove on it by add i ng a simple contrast curve.
3.
In the Laye rs palette, double-click the blending options button to the right of the fi l te r name. In the Blending Options d ialog box, open the Mode menu and choose Pin Light. This l ightens the image quite a bit and creates better ton a l separation between the steeple and the dark sky (Figure 9-30). Other blend i ng modes that can be usefu l with the Sola rize fi lter and grayscale i mages a re Hard Light and Diffe rence.
Figure 9-29. A simple Curves adjustment will always improve on the default solarization treatment.
Figu1'e 9-28. The m'ults ofapplying the Solarizefilte!: The hij·togram for the solarized image reveals what thefilter is actually doing to the tonal values in the image. r!) Sf) 2.
Add a Curves adj ustment l ayer and c reate a cu rve that increases contrast by raising the h ighl ight section of the curve and lowering the shadow section. Click O K (Figure 9-29 ).
Another way to mod i fy the basic solarize e ffect is to mod i fy the Smart Filte r by changing the opacity or the blending modes. Figure 9-30. Changing the blending mode of the Solarizefilter layer can also create interesting variatiom. Here, we're using the Pin Light blending mode.
Darkroom Special Effects I
333
Custom solarizatio n with Curves
- Channel: ! RCB
:j
You can also brew your own custom solarize effect with more control using a C u rves adj ustment layer. 1.
Bring u p t h e History palette a n d click the opening snapshot at the top of the palette. This will retu r n the i mage back to where i t was when you fi rst opened it.
2.
Add a Curves adjustment layer. I n the Cu rve Display Options at the bottom of the dialog, make sure "Show Amount of' is set to Light. Click i n the center of the cu rve grid to add a control point. Drag down the h ighlight poin t in the upper-right corner and reposition i t in the lower-right corner (F igu re 9-3 1 A).
3.
Now c l ick the curve to p lace two control points on either side that reshape i t i n to the shape of an i nverted V. Use the basel ine to guide you on the left side, and then simply copy the basic positions of the control points on the right side (Figure 9-3 1 B).
This is the cu rve to c reate the basic sola ri zed effect. The shadow and midtone values a re unchanged, but every tonal value above level 1 28 is now i nverted, producing the classic half-positive, h a l f-negative look of sola rization. Experiment with mod i fying the cu rve to see how you can alter the sola ri zation effect. Some examples of d i fferent cu rves and thei r e ffects on the i mage a re shown in Figure 9-32 .
Figul'e 9-32. Custom solar'ization cm·ves.
334
I Chapter 9 : Creative Enhancements
.
Show Clipping
o
� 1 203 I 1)'1/I1'1 0 Show CUppIng
Figure 9-31. To create a solarization curve, drag the highlight point down to the lower-right comer (A), and then add anchor points to create an inverted V shape (8).
Solarization curves with color photos The process for sola ri zing color photos is the same as for black and whi te, with the added twist that in addition to strange tonal inversions, you will have color inversions as well (Figure 9-33). The use of customized cu rves and blending modes, and adj usting the individual red , green, and blue cu rves will also create i n teresting resu l ts.
Figure 9-33. Solarization ofa color image.
(: m
Reticulation Reticulation is the term used to describe the cracking and warping of the thin fil m emulsion layer due to a sudden change i n process temperatures d u ri ng development. Reticulated fi l m typically suffers from increased grain and u neven striations that appear as a tightly woven pattern of tiny squiggles that degrades fine details. B u t the creative muse appears in unlikely moments, and at some poin t someone probably looked closely a t the reticulated fil m and thought, " Hey, that looks k i n d o f cool ! " And so, a permanent, detail -degrad ing, film-damaging mistake was adopted as an a rtistic effect. Traditional reticulation in the wet darkroom is a burn your-bridges, say-a-Hail-Mary, no-turning-back kind of affair because the e ffect is i ntroduced as the fi l m is being developed. Fortunately, Photoshop reticulation is not only much easier to do, but i t's also non-destructive and you can fine-tune the effect after you have applied it, or you can combine it with other effects (Figure 9-34). Let's take a look at how i t works.
Figure 9-34. The basic reticulation e.fftct is not that interesting on its own, but it can be effective when combined with other techniques. Convert for Smart Filters to turn the layer i n to a Smart Object.
2.
Choose Filter--->Blu r--->Gaussian B l u r, and set the rad ius to 30 pixels. Click OK. Now set the blending mode for the bl urred Smart Object layer to Multi ply. This creates an i n te resting d a rkening fade that also has a hint of d i ffusion to i t (Figure 9-54) .
3.
Let's explore a fu rther mod i fi cation to the contrast. Add a Cu rves adj ustment l ayer. Make sure that the grid size is set to 1 0x 1 0. Drag the h ighlight point in the upper-right corner down the right side one grid square. Then d rag the shadow point in the lower-left corner up the left side three grid squares. Drag down s lightly on the lower section of the curve. Click OK (Figure 9-55) .
This last Curves modi fication results i n a n i mage that I S hazy and dark with muddy shadow tones and poor contrast. This is certainly not the ideal photographic i mage, but for some imaging projects, precise tonal perfection may not the best approach for creating a certain mood .
Figure 9-56. Adding some grain and tilting the angle ofthe image suggests some rough weather on the Atlantic crossing.
Figure 9-54. A blurred copy ofthe image set to Multiply darkens and creates a hazy diffusion effect at the same time. Channel:
Matching a M odern I m age to a Vintage Print Let's continue with this same i mage of the Queen Mary, ci rca 2005, and compare it to a photo taken on the Queen Mary almost si xty years earlier. Figure 9-57 shows the photo taken ' in 1 947 as the Queen Mary docked i n New York City a fter a voyage from Southam pton, England. The t h i rd man from the right just below the diagonal w i re is Sean's father, James Duggan, arriving from I reland. In the following exercise, our goal is to get the modern image to match the tones a nd feel of the vintage photo. We have even prepared a file that you can download containing a blank paper background that matches the one in the vintage photo. You can use this as a new background for the modern i mage (Figure 9-58). We'll approach this project by add ressing three main areas: matching color and contrast, match i ng the focus, and finally, matching the texture. �� . ------------------�----------------��--,
Figure 9-55. A Curves adjustment drasticallyflattens the contrast and adds to the murkY atmosphere.
For some final atmospheric fun, we added a fil m grain layer using the 50% G ray layer set to Ove rlay, as described earlier in this chapter. Then we chose I m age--;.Rotate Canvas--;. Arbitrary, and entered a value of9° CCW. The final step was to crop to just i nside the tilted i mage border. This creates the look of an u nderexposed photo, hastily taken as the ship pitched and rolled i n high seas (Figure 9-56). Photographs are all about telling stories. With a l i ttle imagi nation and some creative d igital darkroom tech nique, you can create new stories for your i m ages.
F'hOtDlru.....! on " ', " '3�
.
bo.aJ'd VI f7't{
Figure 9-5 7. The Queen Mary arrives ill New York, November 1 947. ' )"
I
Fading Away: The Distressed Image I
343
val ue to 240. In the v i n tage Queen Mary photo, the darkest shadow values fal l a round 45, so the pu rpose of this adjustment is to create a similar level of shadow density (Figure 9-6 1 ) . Click OK.
".,."",.,...-1'--' " _1"1 .. .. :...,""" " .1'
Figure 9-58. Matching a modem image to a vintage print, before alld ajiel:
Figure 9-59. Adding the photo to the vintage pilato paperfile.
Matching color and contrast 1.
The fi rst step is t o a d d t h e 2005 Queen Mary photo to the vintage photo paper image (use the unaltered i mage, not any versions you have a l ready faded). Use the Move Tool to d rag and d rop the ship photo on top of the old photo paper. Position the ship layer so you can see the typewritten "Photographed on board H .M.S Queen Mary" notice (Figure 9-59).
2.
Use the Rectangular Marquee tool to d raw a selection around the photo of the ship. Create the selection so that i t overlaps onto the background but does not incl ude the typewritten notice at the bottom. Activate the Background layer and use Cmd/Ctrl-J to copy the selection onto a new l ayer. Drag this layer above the ship photo i n the Layers palette and set the blending mode to Color (Figure 9-60) .
3.
344
Click the ship photo (Layer I ) to make it active. Option/Alt-click the adjustment layer button at the bottom of the Layers palette, and choose Levels. I n the New Layer dialog box, turn on the option for "Use prev ious layer to create clipping mask." This will apply the Levels adj ustment only to the previous layer. In the Levels dialog box, set the Output Levels black poi nt value to 45 and the Output Levels white poi nt
I Chapter 9: Creative Enhancements
J
Layers [J:hannel, I Paths I Color paclty: 1100%H FIII:lloo"j.) x
• Laver 2
- IX . fCl
�. CJ
!iJ
Ii
," 1
Figure 9-60. The CoLor bLending mode adds the coLor ofthe paper to the image, but the contrast is stiLL a bit too crisp.
Matching Texture
Chillloel:1 RCB
:1
tnpul �veb:
[
..
ft,
� � � �
5.
The final step to complete the illusion is to add some of the worn paper texture into the modern image. Click the top layer of the paper overlay that is adding the slightly yellowed color to the image. Use Cmd/ Ctrl-J to make a copy of this layer. Double-click its name, and rename it "Texture Overlay." Change the blending mode to Normal, and then choose Image--> Adjustments-->Desaturate to remove all the color values from the layer (Figure 9-63).
6.
Apply Levels directly to the Texture layer by choosing Image-->Adjustments-->Levels. Use the following values for the Input Levels: Shadows, 100; Midtone, 0.50; Highlights, 19 5 (Figure 9-64). Click OK. This adjustment compresses the tonal range of the texture layer and enhances the dark cracks and smudges of the original photo paper. More importantly, it makes much of the texture layer white, which is a neutral color for the blending mode we are about to take advantage of.
lId(kgrouna
1'). 0 51 Ii
Figure 9-61. Matching the tone and contrast ofthe newer image to the contrast levels in the vintage photo.
Matching focus 4.
The color and contrast of the image is matching very well to the photo paper background and to the overall contrast of the actual image taken in 1947. Now it's time to address focus issues. Click Layer I (the ship photo) to make it active, and then choose Filter--> Convert for Smart Filters. Choose Filter-->Blur--> Gaussian Blur, and set the radius to 0.8 pixels to add a very slight blur. The object here is not to actually blur the image, but to take away some of the sharpness created by twenty-fIrst-century camera technology (Figure 9-62).
Figure 9-63. Preparing the texture layer.
� � � (�IJon._.)
/. ,
Figul'e 9-62. A slight blur is added to match the sharpness (or lacl( thereof) in the 1947 photograph.
Figure 9-64. Increasing the contrast on the texture to show the smudges and scratches more.
Fading Away : The Distressed Image I
345
7.
8.
Set the blending mode for the Texture layer to Multiply to reveal the worn texture over the ship image (Figure 9-65). Drag the Texture layer below the color overlay layer. Choose Layer-Create Clipping Mask to add it to the clipping group of the levels layer and the ship photo. Since the photo layer of the ship is the base of the clipping group, the transparent areas for that layer establish the visibility for all the layers that are clipped to it. Adjust the opacity of the Texture layer as needed. We decided to lower it to 80% to make the smudges and cracks a little more understated (Figure 9-66). You could also add a layer mask and hide some of the texture if it interfered with important image areas.
Creating Textures for Use in Photoshop Photoshop can be used to create textures, and, with customization, some of them can be useful, as we saw with the earlier examples on canvas and watercolor paper. However, others end up having the canned look of digitally generated textures. Rather than rely on Photoshop to create textures for you, it's much more interesting to go and find your own, either by photographing interesting textural surfaces you run across, or by collecting your own analog, real-world textures and scanning them. Here are a few examples of things that we have used in our low-fidelity, texture-generation process: •
•
•
•
Blank 120 film and film sleeves, stepped on and scraped across rough concrete A gloriously degraded 35mm film leader, found on a New York sidewalk Paper, stained with coffee grinds, crumpled and left out in the sun for a day (Figure 9-67) Paper, imprinted with inky fingerprints and other smudges
•
Semi-translucent, beautifully scuffed plastic
•
Handmade art paper
•
Aged-stained blank pages from 175-year-old books
Figure 9-65. The cracks and dirty texture from the original paper are added to the new image using the Multiply blending mode.
P'hcIto,�">tI...t on k..ro;t " .. �. ':!"-11' "H'"
Figure
346
9-66. Thefillal
version ofthe image.
I Chapter 9: Creative Enhancements
Figure 9-67. A (,(Jed coffee filterfilled with wet grounds created the stains 011 the paper (top). It was combined with the photograph usillg the Overlay blending mode and the Blend !fsliden-. 'I)
Creating digital textures from all of these was much more fun and interesting than using Photoshop to make textures for us! And the results look real because the textures are real. We've already had a hint of real-world texture in action with the smudged and cracked surface of the photo paper that was used in the faux vintage Queen Mary image. In the next few examples, we'll take a look at how different real world textures add character and depth to the photographs they are blended with.
2.
With the top layer of the roots active, choose Layer�Create Clipping Mask. Since the art paper is already a separate layer, when it becomes a clipping mask for the roots layer, the transparent areas of the art paper layer determine the visible shape of the roots image (Figure 9-69).
Combining Photos with Art Paper For this texture we scanned a sheet of rough-surface art paper with torn edges. Combining it with an image is a simple procedure when blending modes and layer clipping masks are used (Figure 9-68).
Figure 9-69. A clipping mask is used to fit the tree layer to the shape of the art paper faye!:
Figure 9-68. A photograph blended with a ,-heet ofart papa
3.
Change the blending mode for the roots layer to Color Burn. This creates a pretty interesting washed-out blend with the underlying art paper texture (Figure 9-70).
4.
Add a Curves adjustment layer and click OK without making any changes. Set the blending mode for this layer to Multiply to darken the roots layer (Figure 9-71).
1/)
't ch09_finarCpaperO.jpg 't ch09_treeroots.psd 1.
The first step is to bring the two images together into one file. Using the Move Tool, drag and drop the roots image on top of the art paper image.
Figure 9-70. The tree layer I�' Jet to Color Bum.
Fading Away : The Distressed Image I
347
Figure 9-71. An empty Curves adjustment layer set to Multiply is used to darl(etl the tree image.
Blending Textures into Skies When Sean was teaching at the Maine Photographic Workshops one summer, one of his students asked how to combine a texture into the sky of a landscape photo. To help answer her question, he photographed some peeling paint on a wall later that afternoon (Figure 9-72). Since the wall was close to white, he knew this would blend very easily with the sky in the photo of Marshall Point lighthouse, which had been taken the day before (Figure 9-73). This tutorial, which was developed to answer the student's question, is a hybrid between adding a texture and creating a simple collage.
/�
Figure 9-73. Peeling paint combined with a landscape photo. 2.
' most of the worl( in combining the texture with the sky. 348
I Chapter 9: Creative Enhancements
3.
At the bottom of the Layers palette, click the Add Layer Mask button to add a layer mask to the paint layer. Use a soft-edged brush tool and paint with black at 100% opacity to disguise the hard edge at the bottom of the paint layer. You can also touch up the few areas where the peeling paint is visible on top of the lighthouse tower (Figure 9-7 5).
- •.
1
�
� .
Figure 9-76. A
section of the sky si copied to a new layer and set to Multiply.
Figure 9-77. A
layer mas/{ is used to show only the clouds on the skY laye}:
Figure 9-75. A layer mask is used to disguise the hard lower edge of the peeling paint laye}:
Currently missing in the image are the clouds from the original lighthouse image. Although we could certainly use the existing layer mask to show them again, that might also hide some of the peeling paint texture. We'll go about this using another method that will allow us to have our clouds and texture, too. 4.
Make the Background layer active. Use the rectangular marquee tool to create a selection of the sky in the lighthouse photo from the top of the image down to just above the horizon. Copy this to a new layer with Cmd/Ctrl-J. Rename the layer "Sky," drag it up to the top of the layer stack, and change its blending mode to Multiply (Figure 9-76).
5.
Add a layer mask to the Sky layer. With the Brush tool and a large, soft-edged brush, paint with black over the lower portion of the sky, including the area where the lighthouse tower was made too dark by the Multiply blending mode (Figure 9-77).
6.
As a final step, add a Hue/Saturation adjustment layer to the top of the layer stack. Click the Colorize checkbox and set the Hue to 38 and the Saturation to 15 to add a sepia tone. Click OK, and then change the blending mode to Overlay and the Opacity to 80%. This adds a hint of the sepiaesque flavor to the image (Figure 9-78).
Figure 9-78. A final Hue/Saturation layer is set to Overlay to add a touch of sepiaesque.
Blending Textures into Skies I
349
Creative Edge Effects Creative edge effects are the finishing touches you can use to dress up an image with a subtle or bold border. These techniques can be used to draw attention to the image or to extend it beyond the standard boundaries of the rectangular or square photographic format. Some edge effects can be subtle and understated, such as a simple black keyline or stroke to make the image stand out more on paper. Other edge effects are almost works of art in their own right, lending rough and sculpted borders to the photo to suggest filed out negative carriers, brushed-on emulsions, and Polaroid transfers. In this final section of the chapter, we take a look at some ways to add a little something extra to the edge.
A Simple Black Stroke We typically do not add a keyline stroke to our layered master file because the size of the stroke is sometimes influenced by how large of a print is being made. This step is typically applied after a flattened copy of the master file has been made and resized to the final print dimensions. 1.
Choose Layer->Duplicate Layer to make a copy of the Background layer. Click the original Background layer to make it active. Make white the foreground color, and then press Option/Alt-Delete to fill with the foreground color.
2.
Tn order for the keyline stroke to show up, you first have to add some extra canvas around the image. Choose Image->Canvas Size. In the Canvas Size dialog box, open one of the menus that show the unit of measurement and change it to percent. Change the width and height to 1 10%. Make sure the Canvas Extension color is set to white and click OK (Figure 9-79). Canvas Size Current Size: a.OOM --------, Width: 2048 pixels Height: 1365 pixels New Size: 9.68M Wldth:�lpercent
:J
Height:�(percent
:1
o Relative
AnCho,,
� �
Canvas extension color:! White
:fD
Figure 9-79. Adding extra white canvas around the image. 350
I Chapter 9: Creative Enhancements
.\/)
3.
Make the main image layer active and zoom up to 100% (View->Actual Pixels) so you can see the stroke well once it is applied. Click the Layer Style button (the small, italicized fx) at the bottom of the Layers palette, and then choose Stroke.
4.
The default color is red, which, while quite noticeable, is not really useful. Click the color swatch and, in the Color Picker, choose black for the stroke color. Tn the main Layer Style dialog box, set the Position to Inside and the size to 2 pixels. The size is governed by how large a print is being made and how wide you want the stroke to be. Adjust the opacity of the stroke if you don't want it to be totally black, and then click OK (Figure 9-80).
Stroke --------, OK Structure-------, Cancel Size: "'"�--� f2!px ( N.w Styl.... ) PoSillon:� �Preview I Blend Mode: !L.:.No, m .1 Opacity:
::::::=::::;r,;;;� :;! :
a fiOOl "
..
["'-,..
:1
Color:_
Figure 9-80. Adding a thin black keyline stro/{e as a Layer Style.
Filed-out Negative Carrier Many photographers like to print their negatives full frame, with no cropping at all. Apart from taking away the decision of how to crop the shot, it shows that the photographer used the camera viewfinder to determine the crop as the photo was taken. Most photography courses make students shoot full-frame at some point to drill home the importance of seeing the composition in the viewfinder and not relying on cropping afterwards. An interesting method that some photographers use to show that the image has not been cropped is to file out the edges of the metal negative carrier. When the image is printed using an enlarger, this means that the entire film frame is visible, down to the black border around the image. The result is a rough-edged black border that creates a custom built-in frame for the image (Figure 9-81). Some photographers even go so far as to file out the neg carrier so that portions of the film sprocket holes are visible.
Figure 9-82. The filed-out negative carrier custom borde!: ,,') 1.
Create an 8-bit new file that is close to the size, resolution, and orientation you normally like to print at. It doesn't have to be exact, since the alpha channel mask we will create can be scaled up a bit with no significant loss of quality. In this example, we'll create a neg carrier mask to fit a standard 35mm aspect ratio.
2.
Choose the Rectangular Marquee tool. In the Options bar, set the feather value to zero and set to Style to Fixed Aspect Ratio. Enter 3 for the width and 2 for the height. Drag out a rectangular selection that fills most of the empty file, leaving a little border around the edges (Figure 9-83).
Figure 9-81. This portraitfrom a series all the 1991 Gulf War was printed using afiled-out negative carrie}: '} If you still have your old black-and-white darkroom gear stored away somewhere, the best way to get a filed-out carrier look is to either scan an actual filed-out carrier or photograph it on a light box. Another analog idea would be to cut a rectangular opening into a piece of tin or sheet metal in a specific aspect ratio (for example, 3:2 for 35mm), and then file the edges of the opening to create the roughened border. If neither of those alternatives is an option, then there are ways of creating a faux filed-out neg carrier in Photoshop (Figure 9-82).
Figure 9-83. Creating the initial rectangular selection.
Creative Edge Effects I
351
3.
PhDllICopyflO0Q
In the Channels palette, click the Save Selection as Channel button to create an alpha channel from the selection. Choose Select--->Deselect (Figure 9-84).
0 'PhoICOIW
� � cO
!b,lu!e.. •
1:1
ii5l 'iI
�TOf"[dg..
Figure 9-86. The settings for the Photocopy fiLter are combined with the Tom Edges filtn 6.
Figure
9-84.
The saved aLpha channel.
4.
Press D on the keyboard to return to the default colors. Choose Filter--->Sketch--->Torn Edges. The Filter Gallery will open using the default settings for Torn Edges. Configure the settings as follows: Image Balance, 25; Smoothness, 3; Contrast, 10 (Figure 9-85).
5.
At the bottom right of the Filter Gallery dialog, click the new effect layer button (circled in red in Figure 98 5) to add another filter to the effect. By default it will be a duplicate of the current filter. In the center panel in the Sketch folder, click Photocopy. Set the Detail to IS and the Darkness to 6 (Figure 9-86).
If you like the way the smooth outer edge looks, then you can click OK. To roughen up the outer edge a bit more, click the New Effect Layer button one more time, and then choose the Conte Crayon filter from the Sketch folder. Set the Foreground and Background Levels to 12, the Texture to Canvas, the Scaling to 1 50%, and the Relief to 5. The Light direction is not that critical (Figure 9-87). Click OK to apply the combined three-filter effect. Save this file so you can use it again. ConliCrilyon (lOOK)
1'••___-.11
.. t:! llu,hSU"k.. " U DIUDrI "�Sbuh
I""",cu_
� �'�N ·�
FOftgrou.wltvtl
�'kot�L.tvt1
Tutu't:!c-I
I fill . ii2l .
".,.,
'1-0 n, s
��,
.,..
.....,
UOhI:ITOCI
0 .....
:1
� .... �-
ji'Torn[dg..
Figure 9-87. The Conte CrayonfiLter effect is added to the Photocopy and Torn Edges effects.
The key thing to take away from this tutorial is that you can combine multiple filter effects using the Filter Gallery. This allows you to build up an edge
Figure 9-85. The settings for the Torn EdgesfiLtn
effect that's not based on any single filter. Many of the Artistic and Brush Strokes filter groups can be used to cook up a customized edge effect. The ones we are showing here are just to give you an idea of how to get around the kitchen and find the different ingredients.
352
I Chapter 9: Creative Enhancements
7.
Before you import the neg carrier frame into an image, make sure that the pixel dimensions of the neg carrier file are either the same as, or similar to, the photograph.
8.
In the file with the alpha channel active, choose Select-All, then Edit-Copy. Switch over to the image file where you want to use the filed-out neg carrier and choose Edit-Paste. The alpha channel will be pasted down as a new layer (Figure 9- 88).
9.
Choose the Magic Wand from the Tool Palette and click in the center of the neg carrier frame. From the Select menu, choose Modify-Expand, and enter a value of I pixel. This will prevent any white fringes from showing along the inner edge. Press the Delete key to reveal the image underneath. Deselect the selection.
10.
If you need to resize the neg carrier frame, choose View-Full Screen Mode with Menu Bar, and then zoom down so you can see the entire image with plenty of gray around it. Choose Edit-Free Transform, and then drag on one of the corner handles to scale the carrier frame larger to reveal more of the underlying image. Hold down the Shift key if you want to preserve the original aspect ratio of the neg carrier frame. Press the Enter key to complete the transformation or press Escape to cancel the process (Figure 9-89).
Figure 9-89. Resize the neg carrierfimne using the Free Transform command. Hold the Shift key down to comtrain the proportions.
Hand-painted Emulsion Edges In the world of platinum, palladium, and other alternative photo processes, photographers mix their own emulsions and then hand-coat the emulsion onto fine art papers. When the print is developed, the areas outside the boundaries of the image reveal the rough black edges created by the brushstrokes (Figure 9-90). The analog way to recreate this technique is to paint a rough rectangle or square onto a sheet of paper with black poster paint (actually, it doesn't even have to be black paint since you can change it to black in Photoshop). When the paint dries, scan the paper and use that as a custom hand-painted edge. Of course, it's pretty easy to do it in Photoshop, too. I.
Open an image you want to work with. Use Cmd/ Ctrl-J to copy this image onto a new layer.
2.
Select the Background layer to make it active. Make white the foreground color, and use the Option/Alt Delete shortcut to fill with the foreground color.
3.
Choose Image-Canvas Size. Change the units to percent and add 150% to the height and width. Make sure the canvas extension color is set to white. Click OK to add the extra canvas (Figure 9-9 1).
Figure 9-88. The copied neg carrier template is pasted dowlI as a new layer 011 the image file.
Creative Edge Effects \
353
Press D to set the default colors and make black the foreground color. Select the Brush tool. Open the Brush picker and click the brush submenu button to open the menu of additional choices, Near the bottom of this menu, click Thick Heavy Brushes, and then choose Append, Open the Brush Picker and near the bottom of the list is a 104-pixel Rough Round Bristle brush (Figure 9-92), Choose this, and then increase the Master Diameter to a size that will allow you to cover the edges without using too many brushstrokes (our brush was 160 pixels),
4.
)1 '" ·IJ8rush'
,1:1
, 104
Master Diameter
( Use Sample Size ) Hardness:
.��. 111 "
1
Mode'"
1160 px
r Normal
I
�
C:=:J
-
"'"'
,
100
�l 104 "g1
0. �
.
ln4
Rough Round 8ristle
Figure 9-90, A custom edge that creates the Loo/( of photographic emu/sian that has been appLied by hand, L SIJ
[
Canvas Size
Current Size:
New Size:
7.23M Width:
1852
pixels
Height:
1365
pixels
I6.3M Width: Height:
�( percent :1 �( percent :I
o Relative Anchor
l�mB /1 '..
Canvas extension color:
( Background
:10
� ==�--� Normal : l Opac.ily: [0!E Flll
layer
r:-£
1
Backgrouna
Figure
354
9-91,
Adding extra CCl1wa>'to the image,
I Chapter 9: Creative Enhancements
Figure
9-92,
-0-
Selecting one of the Thiel( Heavy brushe>',
5.
The Background should still be the active layer. Start brushing over it so that the black brushstrokes extend beyond the edges of the image. Continue brushing in the same direction until you have covered all the edges. Don't worry about covering the interior areas since those can be easily filled with black using a selection (Figure 9-93).
[f you just want to have roughened, brushed on edges behind the image layer, then you are done. This is how it would look if a large negative had been contact-printed on hand coated paper. If you want to explore adding a roughened edge to the image itself, read on. 7.
Turn the image layer visibility back on and click the layer to make it active. Click the Add Layer Mask button to add a layer mask to this layer. With the Brush tool active, open the Brush Picker and change the Master Diameter to 50 pixels. Paint with black on the layer mask over the edges of the image to mask the edges using the rough brushstrokes (Figure 9-9 5). _ x
�__J� I�;;���==��
Figure 9-93. The completed bruslwro/(es defining the edges underneath the image faYe!: 6.
When you have finished covering the edges with brushstrokes, turn the visibility off for the image layer by clicking the eye icon. Make a rectangular selection of the area that is still white. With black as the foreground color, press Option/ Alt-Delete to fill the selection with black. Choose Select-Deselect (Figure 9-94).
Figure 9-95. Adding the rough brushstro/(es non-destructively to the edges of the image using a layer mask.
The rough round bristle brush tip that we used for this example is only one of many potential Photoshop brushes that can work for this type of technique. Explore some of the other brush sets and see which brushes work well with this technique on your images.
Figure
9-94.
FiLling the interior with b/ac/(.
Creative Edge Effects I
355
Enhancing the Visual Story Good photographs tell a story and make the viewer want to know more about the scene or ideas portrayed within the frame of the image, or generate curiosity about what happens next or what transpired just before the shutter button was pressed. Sometimes the creative path you decide on will include enhancing the photograph in various ways to create a mood, add emphasis to certain elements within the image, or hint at intangible qualities such as memory, dreams, or emotion.
As we have seen in this chapter, some of these techniques can suggest the appearance created by traditional photographic filters, film types, and darkroom processes. These can be subtle-a slight accent or a little added seasoning for a photo-or they can transform an image into something very different from the original source material. But the image and the story it tells should always come first. Create the story first, and then build on it, enhance it. In a way, image enhancements are like the well-chosen words and finely crafted sentences that a writer uses to add depth, detail, and nuance to a story. They help to create a richer narrative experience for the reader. In a photograph creative enhancements can be used to accomplish the same goal for the visual story presented within the image.
Figul'e 9-96. "The Open Window" has been seasoned with some sepiaesque dijjilSion toning.
356
I Chapter 9: Creative Enhancements
"' Sf)
Figure 9-97. To create a nostalgic mood a11d suggest memories ofa dista11t summel; "A Measure of Longing" has been enhanced with diffusio11, selective blurring and toning. � Sf)
Enhancing the Visual Story I
357
Enhancing Focus Photographers love to study lens specification sheets, read technical articles, and squint at shots of test targets to determine lens sharpness, resolving power, edge-to-edge performance, and possible contrast or distortion issues. It takes a lot more than good equipment to take a strong image, and even when using the sharpest lenses your images may not be in true focus that expresses the photographer's visual and conceptual intent through appropriate use of depth of field, sharpness, exposure, and composition. In the traditional darkroom, the ability to refine focus and sharpness, and remove distractions was difficult if not impossible to do. In the digital darkroom your files are the raw ingredients, and with enough time and layers you can remove, conceal, or reduce distractions, adjust depth of field, and sharpen your images to allow the viewer to see what you saw when pressing the shutter. This chapter addresses the skills and techniques to: •
Remove Distractions
•
Improve Backgrounds
•
Sharpen with Finesse
•
Add Valuable Noise
From a digital workflow pointof view, all focus enhancement improvements are best done in Adobe Photoshop with well-exposed and well-processed images. Although Adobe Camera Raw and Photoshop Lightroom have good input sharpening capabilities, carefully enhancing focus requires working selectively with layers and layer masks. Finally, just because an image isn't sharp or in perfect focus doesn't make it an automatic candidate for rejection: both of us love to experiment with pinhole cameras and toy lenses to create quirky, unsharp images that often better express our perception of the scene than a tack-sharp image ever could.
Chapter 10
Remove Distractions Distractions come in many shapes and sizes, and the simplest way to find them is to defocus your eye, as illustrated in Figure 10-1. Notice how your eye is drawn to the white clump in the lower-middle part of the image and how, after it has been removed, your eye now remains on the subject of the image, as seen in Figure 10-2. Our eyes are attracted to color, contrast, and light, and distractions often intrude in on the edge of the frame or are high-contrast elements such as light against dark shapes or dark against white shapes. Classic distractions include: background elements that are uncomfortably close to the subject such as tree branches protruding from a head, specular highlights, and edge elements that, in the rush of taking a photograph, were unavoidable. Figure 10-2. After removing the scrap of white papel; the image focuses on the tu·ming pages. For some photographers, removing distractions such as a soda can in a landscape is unethical and does not render the scene as it originally was. For us, photography is about creating the best image, and we can either move the soda can before pressing the shutter button or afterwards with Photoshop. In most cases, we'll opt for picking up the can, throwing it into a trash bin, and then taking the picture. Photoshop is a continuation of the photographic process, and many times it isn't until we're back in front of the computer that some distractions become glaringly obvious. Each photographer needs to think about how they interact with the scene. Is it OK to bend a grass stem down so that it doesn't interfere with the frame, or is that manipulation? Is it all right to clone out a bit of trash, or does that change the image?
Please see the Cropping and Straightening Images section in Chapter 4, "File Preparation" for valuable information on the importance of cropping to refine image composition.
Essential Clean up
Figure 10-1. Defocusing yow' eyes as you look at a print or afile emphasizes the distractions. () KIl
360
I Chapter 1 0: Enhancing Focus
We don't live in a perfectly pristine world; people still litter and, as the before image in Figure 10-3 shows, animals litter, too. The white flecks on the red carpet are very distracting, and by removing them, the visual focus of the image falls onto the eagle, as seen in the after image. The most common tools for cleaning up an image are the Clone Stamp and Healing Brush tools.
Figure 10-3. Removing the out-offocus white specl(s allows the visualfocus ofthe image to retum to the eagle
.
When cleaning up i mage backgrounds with l i ttle detail, the Patch tool is also excel lent at quickly remov i ng d i rt, d ust, and small pieces of litter.
'11 ch10_eagle.jpg 1.
I f working with a si ngle-layer i mage, d u plicate the background layer, or, if working with a multilayer image, click the topmost layer, and then press Cmd Option-ShiftlCtrl-Alt-Sh i ft+E to merge all working layers up.
2.
Select the Patch tool and make sure that the options are set to Source and that Transparency is not selected.
3.
Generously circle the d istraction with the Patch tool, as seen in Figure 1 0-4, and then d rag the selection to a clean area in the image. For best results, choose a clean area near the d i rty a rea that matches the tonality and texture of the a rea to be cleaned up.
4.
Move onto the next piece of d i r t and repeat the ci rcle and d rag process for each piece of d i rt. We fi nd that using the Patch tool is m uc h faster than cloning or heali ng the i nd ividual spots.
/
"'r
..
' f.J
.
..;
i'
,/
Figure 10-4. Circle the offending area and drag it to clean at'eas to quickly remove spots, specks, and distractions.
Remove Distractions I
361
In addition to removing trash, take a moment to look at your images to double-check if out-of-focus or d istant birds, bothersome electrical w i res, distracting highl ights or reflections, or flyaway hai r should be removed with a combination of cloning, heali ng, and patching.
JEJ ch10_childs-play.jpg 1.
If you are working with a single-layer image, add a new layer, or if working with m ultilayer images, click the topmost layer, and then add a new layer.
2.
Activate the Clone Stamp tool and make s u re that "Current & Below" is active in the sam ple option. This i nstructs Photoshop to sam ple all visible layers and place the new i n formation onto the new active layer. In earlier versions of Photos hop, this is called "Use All Layers," and i t was very i m portant to have the clone or heal layer above all the adjustment layers to avoid having the repairs being adj usted twice.
Improving Backgrounds The primary methods to improve backgrounds are clean up, darken, blur, or replace-with the first three being more straightforward than a complete background replacement, which is addressed thoroughly in Katrin's Photoshop Masking & Compositing book. Of course, many of these attributes can be controlled i n the camera by using a wider aperture to soften the background or using a flash to overpower the sun and darken the background, but admittedly we often see these possibilities m uch later i n our images.
I n Photoshop CS3 you can now instruct the Clone and Healing tools to ignore adjustment layers
Rem oving unwanted elements In the rush of the moment, it's easy to lose sight of the entire image composition, as Katrin did i n Figure 10-5, in which she was concentrating on the l ittle boy playi ng i n the puddle and didn't notice that the j ungle gym in the background was going right t hrough his head. In the after image, you can see how much cleaner the image is, and, more i m portantly, that the heavy contrast of the i ron in relationship to the l ittle boy has been removed, allowing you to see what Katrin saw on a late wi nter day.
Avoid the temptation of removing the biggest distraction first. Many times cleaning up the smaller distractions creates more suitable replacement material that you can use to conceal the larger areas.
when working with multi-layer images by clicking on the button with the adjustment layer icon and a slash though it. When the button is dark, cloning and healing will not include adjustment layers and the repair layer can be placed anywhere in the layer stack.
3.
Clone over the smaller pieces of trash from the lawn in the upper-right corner, as seen i n Figure 1 0-6.
4.
To conceal the j ungle gym, continue cloning the j u ngle gym and don't be afraid of getting too close to the boy's face. In fact, i t is better to add a bit too m uch information than too l ittle. To clean up any u nneeded cloned grass, use the Eraser tool with 50 percent Hardness to remove any telltale grass from his face. In most of our creative i mage enhancement work, we
Figure 1 0-5. By removing the heavy iron poles in the background, the image concentrates on the playing boy.
362
I Chapter 1 0 : Enhanci ng Focus
« 1\/'"
both opt to use the layer mask to hide and reveal i mage i nformation-but when doing cleaning up, ca reful use of the Eraser can be q uicker and j ust as e ffective.
Figu1'e 10-7. Selecting an area to be used to conceal distractions. 6.
Use the Move tool to move the piece into position and, in this case, use the Edit-Free Transform to scale and size the piece as needed. B y scal ing the top down and skewing the left side down, the lower shadow aligns perfectly with the existing shadow, as seen in Figure 1 0-8.
Figure 10-6. By cleaning up the simpler areasfirst, you can "get to I(now" thefile and plan the strategy to taC/(le the more challenging areas. c, "F Depending on your sensibilities, removi ng the vertical bar that was intersecting with the boy's head may be all that is needed to i m prove the image, or you may be more ambitious and want to remove the entire j ungle gym. To remove larger pieces, we find i t more effective to copy good i mage information and paste it onto a separate layer to conceal unwanted i mage elements. 5.
To remove larger distractions, select suitable replacement material and, when work i ng with multi layer images, use Edit-Copy Merged to copy all of the v isible image i n formation. I n this example, use the Marquee tool with a 2-pixel feather to select the area ad jacent to the vertical pole (Figu re 1 0-7) followed by Edit-Copy Merged and Edit Paste. Figure 10-8. Use the Tramform controls to bend the new layer into place. 7.
Conti nue selecting and usi ng Ed it-Copy Merge and Ed it-Paste to quilt additional pieces of i n formation over the j u ngle gym. I n the process, the replacement pieces may look clumpy, but use the copy paste technique to build up initial i mage i n formation that will be refi ned in the final step.
Remove Distractions I
363
8.
After building up the replacement material, turn off the background (Figure 10-9) and choose Layer Merge Visible to combine all the layers that contain new background material.
Figure 10-]0. Darl(ening the edges ofthe image emphasizes the little boy.
To clone or to heal?
Figm'e 10-9. Me,ging the concealing layers into one. 9.
364
Use the Healing brush and Clone Stamp tools to paint over obvious seams, and use the Eraser tool to erase away u n needed i mage remnants.
10.
Darkening down the background is an optional final step to focus the viewer's eye on the l i ttle boy. Add a Curves adj ustment layer and change the blend mode to Multiply.
1 1.
Choose I mage-Adj ustments-Invert and use a White, Transparent gradient on the layer mask to pull from each corner to the center of the image. Use a la rge black brush to conceal any darkening that fel l on the l i ttle boy. Lastly, adjust the layer opacity to darken down the background ever so slightly (in this case, we used 50 percent), as seen in Figure 10- 10.
I Chapter
1 0:
Enhanci ng Focus
The Clone Stamp tool rep l icates pixels and is best used to build up new i n fo rmation w he n concealing large distractions. The Healing brush is a texture replacement tool, best used for smaller details and as a refinement tool after cloning or past i ng i n i m age information. When using the Heali ng B rush, avoid areas of h igh contrast to reduce the potential of add i ng unwanted smudging. To create real istic replacement a reas, avoid repeating patterns by Option!Alt-clicking different source areas in the i mage. Also, be s u re to remove shadows of image elements you have removed.
Darkening backgrounds Darkening down background elements is an effective method to reduce distractions and, in many cases, darkening is combin ed with blurri ng tec h niques to deemphasize the background even more effectively.
To blur or to sharpen? Whenever you blur a n i mage a rea to reduce v isual importance, you can accentuate the blurred e ffect by sharpening the i mage's opposite areas. The sharper areas will make the blu rry areas look softer, and the soft areas will make the sharp a reas look even sharper.
Figure 10-11. Simplifying the backgroundfocuses the eye on the subject.
In the example seen in Figure 1 0-1 1, the colorful splotches in the background distract from the bird handler's bracelet and bag. As you can see in the after figure, removing and da rkening the distractions makes the i mage more coherent. After improving the background, Katrin also sharpened the silver bracelet and leather details to accen tuate the texture.
'1l ch10_hawk-handler.jpg
/..,
Remove Distractions I
365
1.
Add a new layer and use a 50-percent hard ness Clone Stamp tool with "Current & Below" active, and clone over the out-of-focus red sleeve of a person in the background, as seen in Figure 10- 12.
Figure 10-13. Selecting the bacl(grotmd is thefirst step in darl(ening it. 3.
Add a Cu rves adj ustment layer and, as Jeff G reene l ikes to say, the selection will "auto-magically" be transferred to the layer mask - so that only the selected areas will be affected. Darken down the selected area, as seen in Figure 10- 14. After c l icking OK, use a small black or white b rush on the Curves a d j ustment layer mask to refine the edge of the d a rkening. Painting with white will add more darkening, and painting with black will remove darkeni ng. Curves
! Custom Channel: ( RGB
,:II!¥J � �
Preset:
[:1
"
Smooth
,
� ( Options... )
Figure 10-12. Cloning over distractions to conceal them.
1!1 PrevIew 2.
To d a rken the lower area, use a l-pixel feathered Lasso to select the lower area, as seen in Figure 10- 13.
.....
Output:
.
____I __
Input:
0
1/1 .11/1
Show Clippi ng
G Curve Dlsplav Options
Figure 10-14. Darl(ening down the background with Curves.
366
I Chapter 10: Enhanci ng Focus
Blur This! Photoshop features many blurring filters, with some offering more control than others, while some blur the entire image and do not blur i m age edges. I n Photoshop CS3, choose Filter�Convert for Smart Filters to non destructively experiment, adj ust, and refine the filter settings. To e fficiently experi ment with the blur filters with versions of Photos hop prior to CS3, try the filter on a smaller section of the image by:
•
•
•
1.
Selecting a section of t h e i mage w i t h t h e Marquee tool (no feather required).
2.
Using the desired B l u r filter and settings.
3.
After applying the filter, studyi ng the effect and i f it needs adj ustment, undoing t h e effect a n d trying again with a new B l u r filter setting.
•
•
A fter determining the desired filter settings, making s u re to undo the change to the selected a rea. Select�Deselect and use Cmd/Ctr l + F to apply the last used filter to the entire image.
4.
Understanding what each filter does increases your options to create unique i mages. Filter�Blur •
•
•
•
•
•
•
Average: Analyzes the color values in a selection or layer and creates one "average" value. Not very useful for focus enhancement, and best used for colorcast removal. Blur: Smoothes transitions by averaging the pixels next to the hard edges of defined l ines and shaded areas. Best used for m ild noise reduction.
Gaussian: A fast and adj ustable blur filter that adds a hazy effect. Best used on layers and channel masks to create feathered transitions. Lens: Used to mimic the appearance of a wider aperture. The Lens Blur filter works best on backgrounds that have natural h ighlights as the filter over-brightens them very n icely as a wide open lens would. Sadly, the Lens Blur fil te r does not work on a Smart F i l te r layer.
Radial: Adds zooms or spin-effect motion blu rs with higher settings c reati ng a more pronounced effect. Drag the grid in the Motion B l u r interface to define the blur origin. Smart: Blurs detail without blurring edges. Shape: Uses a shape from the custom shape presets to for m the blur. H igher settings enlarge the shape the fil te r uses to c reate a more blu r red effect. Circular shapes can c reate inte resting lens-like e ffects. Smart: A very precise and controllable blur filter that sadly doesn't work i n 16-bit and can be painfully slow to use. The Radius value determines the size of the a rea sea rched for dissi m i la r pixels. The Th reshold value determines how dissimilar the pixels m ust be before they a re affected. To see an initial e ffect you need to increase both the Radius and the T h reshold sliders. Surface: B l ur s while preserving edges. The Radius slider controls the strength, and the Threshold slider controls how m uc h tonal values have to d i ffer to be i ncluded in the blur. Higher val ues increase the effect. U seful for removing noise and for creating hazy e ffects while maintain i ng edges.
Filter�Noise •
Blur More: Three to fou r times stronger than the Blur filter. We use B l u r and B l u r More only for quick noise reduction, as these filters do not offer any control or adj ustments. Box: Blurs an i m age based on the average color of adjacent pixels. H igher settings i ncrease the blurring e ffect.
Motion: Smears the i mage as when using a slow shutter speed. Works best on i m age elements that a re isolated on transparent layers.
Median: Averages tonal and color values while maintaining edges. The filter searches for pixels of similar brightness, discards pixels that d i ffer too much from adjacent pixels, and replaces them with the median brightness value of the searched pixels. Median is useful for reducing the effect of motion and noise reduction.
Filter�Other •
•
Custom: Used to brighten, darken, blur, and sharpen images. For additional i n formation, please see the Custom F i l te r sidebar later i n this chapter. H igh Pass: Works on d u pl icate or merged layers. When used in conjunction with Overlay, Soft Light, or H a rd Light blending modes, it is an effective sharpeni ng fil ter. To create a beautifu l blur, u s e Cmd/Ctrl +1 t o i n v e r t the H igh Pass layer, which unsharpens the sharpening e ffect wonderfully.
Remove Distractions I
367
4.
To sharpen a reas selectively, merge all layers up with Cmd-Option-Shi ftlCtrl-Alt-Sh i ft+E. In Photoshop CS3, select Filter-Convert for Smart Filter. ( I f you're working with an early version of Photos hop, skip the Convert for Smart Filte r step.)
5.
Choose Filter-Sharpen-Smart Sha rpen and start with a moderate Amount of 150 percent with a Radius setting of 1.0, and make s u re to select Remove: Lens Blur.
6.
These settings sharpen the entire image, which is not desired. Click the Smart Filte r layer mask and Cmdl Ctrl + I to invert the white mask to black. For earlier versions of Photos hop, add a layer mask and i nvert it.
7.
Use a soft-edged white brush and paint over the silver bracelet and the leather details of the bag and glove to paint the sharpening onto the select detai ls, as seen in F igure 10-15.
When photographing on location and on the move, it can be d i fficult to control every image element. Darkening the background as described is an e ffective technique to show the viewer what you saw. Adjusting Depth of Field Control l i ng what is and what is not in focus is an essential component of effective i mage composition that is primarily controlled with the lens aperture. La rger apertures (smaller numbers) create a shallower depth of field, as seen in Figure 10- 16, in which the young girl is i n focus and the unimportant background is out of focus. You can m i mic this effect with the Lens Blur fi lter to create the effect seen in Figure 10- 17, which focuses your attention on the handsome man in the image. Due to fundamental lens construction and size relationships, creating shallow depth of field effects is easier done with both longer lenses and larger sensors or film formats. The depth of field decreases as: •
•
•
A lens is opened up-that is, the flnumbe r is decreased. The focal length of the lens increases. Wide-angle lenses have a greater i nherent depth of field than telephoto lenses. As camera-to-subject distance inc reases. The closer you are to the subject, the shal lower the depth of field is. As magnification increases, depth of field decreases.
Figure 10-15. Selectively sha/penil1g the silverjewelry and leather details gives them greater visual importance as the close-ups reveals.
368
I Chapter 1 0 : En hancing Focus
Figure 10-16. Using a !cllge aperture created the shat/ow depth of field \/ « .
.
To control blurring effects, work on duplicate layers, which allows you to use higher filter settings and the ability to refine the filter effect using the layer opacity and blending modes.
Figure 1 0-17. A "hat/ower depth offield can be created with the Lel1J Blurfiltel: f...J
JEJ ch10_castle-visit.jpg 1.
To create this e ffect, you start by creating a depth map that defines the parts of the image that will be in focus and out of focus. The depth map is an alpha chan nel that controls what is in and out of focus, with the dark o r black areas remaining in focus and the white or l ighter a reas being out of focus.
In the Channels palette click the Create New Channel button and make s u re that the channel mask is white. If yours is black, use Cmd/Ctrl + [ to i nvert it. Click the View col umn next to the RGB icon to see the image through the mask, as seen in Figure 10-18.
Remove Distractions I
369
3.
To make sure that the subject's face remains i n focus, in the alpha channel use a soft-edged black brush and paint over his face, as seen i n Figure 10-20. Soften the alpha mask with a Gaussian Blur of 10.
Figure 1 0-18. C1·eating and viewing the alpha channeL mask. 2.
Reset the foreground and background colors by tapping D. Use the gradient tool with the reflected gradient setting (ci rcled in Figure 10-19), and pull from the man's face to the top of the first stone wal l , as seen in Figure 10-19.
Figure 1 0-20. Painting over the mew'sface insures that he will remain infocus. Reverse
Figure 10-19. The reflected gradient allows you to dmw in the line offocus.
370
I Chapter 10: Enhancing Focus
!!I Dither !!I Transparency
4.
I n the Channels palette, click the View column next to RGB to turn off the translucent view and return to the Layers palette, and then click the Background layer. Duplicate the Background layer, and select Filter Blur-Lens Blur.
5.
Near the top of the Lens Blur control column is a Depth Map Source pull-down menu. Choose the Alpha channel you created, which is simpl y called Alpha 1.
6.
Allow the filter to update, and study the e ffect before adj usting the sliders. I f too m uch is being blurred, adjust the Radius.
7.
Adjusting the Blur Focal Distance s h i fts which area i n t h e image is i n focus a n d is best used t o refine what is i n and out of focus. Higher Blur Focal Distance raises the focus area, while lesser val ues lower what is i n focus i n relationship t o the edge o f the frame, a s seen in Figure 10-2 1.
the expensive lens, as seen i n the comparison featured in Figure 10-24. T his technique is most e ffective with i mages that were photographed from an elevated poin t of view and can be especially fun when people or cars are in the image.
'
1:;:"-
�--
Figure 10-22. Photographed with a perspective control lens with settings to exaggerate the size l'elationshipsforces the background to recede. il \ John ,\/{ lUlU,';'
Figure 10-21. Refining the position ofthe Lens Blur with the Blw' Focal Distance slidel: I n case you're not satisfied with the Lens Blur effect, you can either add a layer mask and apply a black brush to the blurred layer to paint out areas that are too blurred, or exit out of the Lens Blur filter, adj ust the alpha mask, and then reenter the Lens Blur d ialog box. Blurring i mage backgrounds can be subtle or d ramatic; as long as you work on duplicate layers, you can always either refine the effect or return to the original i mage and start over. Creative depth of field effects A very popular e ffect is to c reate a very shallovv depth of field to make the in-focus areas look like a child's toy model as seen in Figure 10-22 and Figure 1 0-23, which were photographed with a Canon perspective control lens with tilt and shift movements. You can create a similar effect without
Figure 10-23. Photographed with a perspective control lens to not be in pelfect perspective changes the size l-e/atiollships ill the image. (i' John \/( /1I(1}.,ll
Remove Distractions I
371
Figure 10-24. Aggressively bltl1'ring the back- andforeground makes the row ofhousej' look lil{e a toy model.
'1l ch10_model-village.jpg 1.
Tn the Chan nels palette click the "Create new chan nel" button. Click the View column next to the RGB icon to see the image through the mask.
4.
Nea r the top of the Lens Blur control colum n is a Depth Map Source pull-down menu. Choose the Alpha chan nel you created.
5.
Before adjusting any additional sliders, allow the filter to update and study the e ffect. Tn case too much of the houses or your own subject i s being blu rred, i t is best to cancel and return to the Alpha channel and paint i n some more black t o protect more o f the i mage. Then select Filter�Blur�Lens Blur to conti nue.
6.
Adjusting the Specula r Highlights slide rs increases the highlight blur effect. Change the Th reshold slider fi rst to be able to see the effect that the brightness slider is having, as seen in Figure 10-26. Lower and lower Threshold settings instruct the Lens Blu r filte r to consider more tonal values in the brighteni ng process. Higher brightness settings push the h ighl ights towards pure white.
Use a large, soft-edged black brush to paint over the row of houses, as seen in Figure 10-25.
2.
1-./'
c=
Cf::o---
]
� ··l� -��1 I . 1=--' ._r _
"- .
-
..
In-
c--
.-
Figure 10-25. Painting over the sections in the image that should remain in focus. 3.
372
I
Uncheck the RGB View column, and then click the RGB icon to return to the full color view. Click the Layer palette, duplicate the background layer, and select Filter�Blu r�Lens Blur.
Chapter 1 0 :
Enhancing Focus
Figure 10-26. Adjusting the Specular Highlights sliders mimics using a wide aperture, as highlights blur out more strongly than darl(er tones.
Experimenting with the Lens Blur fi lter can often make up for limitations of equipment or camera setti ngs that you couldn't control while shooting.
JEJ ch10_pecgraves.jpg 1.
Duplicate the background layer and use the Lens Blur filter to aggressively defocus the image.
Blurring backgrounds
2.
Click the layer mask button to add a white layer mask.
Combi ning a shallow depth of field while sha rpen ing the subject matter effectively makes the subject look even sharper than it really is, which also draws more attention to what you took the picture of in the first place. In the example seen in Figure 1 0-27, Katrin was intrigued by the pet cemetery in the P residio Park by San Francisco Bay, but upon viewing the i mage on the monitor, she realized she needed to emphasize the grave memorial as seen in the after version in Figure 1 0-27. By blurring the background and foreground and sharpen ing and brighten ing the grave, the image is closer to what Katrin saw in her mind's eye when she pressed the shutter.
3.
To create a shallow depth of field effect, use a reflected gradient on the layer mask and pull from the top of the picket fence up to the top of the tombstone. Since the reflected gradient tool effect is not perfectly predictable, you may need to try a few d rags to create the desi red e ffect.
4.
Use a large, soft-edged black brush to pai n t over the top of tombstones on the layer mask to conceal any u ndesired softeni ng, as seen i n Figure 10-28.
Figure 10-27. Combining blurring and sha,pelling emphasize>' the graues.
r:J
Remove Distractions I
373
7.
Use the Smart Sharpen filter with a h igh Amount of 1 25 and a low Rad i us value of 0.8, and Remove Lens Blur, as seen in Figure 1 0-29. To accentuate the texture of the picket fence and tombstones, check More Accurate, which i n te r nally i nstructs Photoshop to run the filte r twice with lower settings to build up the sharpening.
�
� �Preview
@ Basic
0 Advanced
= :J � 8
u',,D.::;. "':;: ' _ Settings: !...: _
Amount:fU5I"
Radius:
[D.8J pi)lels
Remove:! lens Blur
Ang"
m
r--l . -
6i1 More Accurate
Figure 10-29. High Amount and low Radius values are best usedfor low-noise digital camera files.
Figure 1 0-28. Painting on the layer mask with black conceals the softening effect from impacting the tombstones.
374
5.
To strengthen the softening effect, duplicate the blurred layer, and change the blending mode to Darker Colors.
6.
With the topmost layer active, press Cmd-Option S h i ftlCtrl-Alt-Shi ft+E to merge all layers up, and then select Filter-'>Convert for Smart Filters.
I Chapter 1 0 : En hanc i ng FocLis
8.
To balance the sharpened i mage i n formation with the blurred image a reas, press Cmd-Option-ShiftlCtrl Alt-Shift, and d rag the layer mask from the blurred layer i n to the layer mask slot next to the sharpened layer icon. This t ransfers, duplicates, and i nverts the layer mask in perfect registration, as seen in Figure 1 0-30.
9.
Brightening the sharper a rea will d raw more attention to it, and this is easily achieved by changing the sharpened layer blending mode to Lighten to c reate the final image.
Darkening and bl urring backgrounds Our eyes move from dark to light, soft to sharp, and less color to more color. Knowing this enables you to recompose images that appeal to how our perceptual system works naturally. Whenever we blur a part o f an i mage, we make sure to sharpen the opposite area, and when we l ighten a section of the image, we darken i ts coun terpart section of the i mage. By delicately balancing sharpness and tonality, you can guide the viewer's eye to see the i mage of your intent. Take a look at the before i mage i n Figure 1 O-3 1 -a qui rky portrait of a cu rious sheep-and now compare it to the after i mage. Which i mage do you think is more effecti ve and best concentrates your eye on the subj ect of the i mage? Applying both blurring and darkening a llows the background to recede even fu rther. To replicate this e ffect or to apply it to your own i mages, you'll use a combination of image adj ustment layers, sharpeni ng and blurring filters, and some slight retouching to conceal distracting h ighlights and unnecessary details. Tn the following example, we'll use the Smart Filter feature that does not support the Lens Blur fil ter. However, Smart Filte rs do support the Shape B l u r, enabling you to a d d beautiful and creative blurs as described here.
When experimenting with sharpening or blurring effects, it is essential to work on a duplicate layer. If Smart Filters support the desired filter, we recommend taking advantage of this incredibly flexible
and
non-destructive
Photoshop
CS3
feature.
JeJ ch10_sheep.jpg 1.
Start by duplicating the background layer and converting it to a Smart Fi lter layer.
2.
Choose Filter-- B l u r--Smart B l u r, and click a shape to see its effect using a Rad i us setti ng of 15, with h igher Radius settings creating more blur.
Figure 10-30. Sharpening and blurring opposite image areas enhances the effect of both.
Remove Distractions I
375
Figw'e 10-31. Judicious use ofthe SUifilce Blurfilter and distraction removal emphasizes the sheep's curious gaze.
376
3.
To best mimic lens d i ffusion we've d iscovered that using geometric shapes with i n ternal spaces, such as the circles or squares, yield pleasing results, as seen in Figure 10-32. To experiment with more shapes, use the small , black triangle flyout menu to load additional shapes including "frames" and "shapes"; make sure to click "Append" to add the shapes to the Shape palette.
4.
Add a layer mask to the bl urred layer and, using the Black, White gradient, start the gradient slightly above the sheep's rum p and drag down to the top of its head. The top of the layer mask is white, which allows the blur to show through, and the bottom part is black, which blocks the sheep from being blurred, as seen in Figure 10-33.
I Chapter
1 0:
Enhancing Focus
r;)
johll .\JrJIIt"sh
5.
To darken the blurred area, add a Curves adj ustment layer and click OK. Change the layer blending mode to Multiply and reduce the opacity to 50 percent, which darkens the entire image. To transfer the layer mask from the blurred layer to the darkeni ng layer press Option/Alt as you drag the layer mask from the bl urred layer to the Curves layer. Click Yes to the Replace Layer Mask? pop-up wi ndow to create the image seen in Figure 10-34.
6.
To emphasize the sheep, apply subtle sharpe n i ng and lightening to draw the v iewer's eye towards its face. To sharpen j ust the sheep, duplicate the blurred Smart Object sheep layer, and drag the Shape Blur smart filter into the small trash can in the Layers palette.
� � e1Prevlew Shape:
D -�"-OfJ"'.\l! ��lBl8< +ia.."
Figure 10-32. Changing the shape influences the overall effect and increasing the Radius number strengthens the bhtl:
Figure 10-34. Darkening the blurred areasforces them further into the distance. 7.
Click the layer mask, and press Cmd/Ctrl + I to i nvert the layer mask. So far i t doesn't look as i f anything has changed, but by choosing Filter-Sharpen-Smart Sharpen, and using very mild settings of 75 percent, 0.8 pixels, and then removing the Lens Blur, the details are made crisper wi thout looking over-sharpened, as seen in Figure 1 0-35.
8.
To l ighten the sharpened sections, add a Cu rves adj ustment layer and click O K. Change the layer blending mode to Screen, and reduce the layer opacity to 50 percent. Option! A l t-c1ick the thin l i n e in the Layers palette between the sharp layer and the Curves layer to clip the Curves layer with the sharper layer, as seen i n Figure 1 0-36. A d j ust the l ightening layer opacity to taste-in this example we needed only 20 percent to create the desired effect.
Figure 10-33. The layer mask controls where the blur is visible.
Remove Distractions I
377
Figure 10-35. Sharpening the lower section ofthe image accentuates the blurring effect ofthe background.
Figure 10-36. Clipping the Curve adjustment layer to the shmpened {ayer limits its lightening effect to the sharpened image areas.
Figure 10-37. Concealing distractions in the bacl(ground encourage.; your eye to remain on the image subject.
378
I Chapter
1 0:
En hancing Focus
9.
Add a new layer at the top of the laye r stack. Use the Clone Stamp and Healing brush set to "Current & Below" to edit out distracting elements such as the highlight on the tree, the sheep in the background, and the ear tag on the primary sheep to create the fi nal i mage, as seen in Figure 1 0-37.
Tinting, darkening, and blurring In addition to blurring and da rkening, you can use a color tint to create a mood, as seen i n Figure 1 0-38, i n which Katrin created a cool "day for n ight" e ffect to better express the sense of loss felt at a child's grave.
'1J ch10_childs_grave.jpg 1.
To create the da rk-blue mood, add a Photo F ilter adjustment laye r, and use the Cooling Filte r (80) at 75percent density. Of course, with your i mages, the color and density decision depends on the i mage.
2.
To enhance the dark-blue effect, change the layer blending mode to Multiply, as seen in Figure 1 0-39.
3.
Use a soft-edged, black brush on the Photo Filte r layer mask and roughly paint out the primary flowers on the grave. This mask doesn't need to be super-precise as the bl ueness can affect some of the flowers, too.
4.
To create an even more solitary mood , Katrin opted to use an additional Cu rves layer, which i l lustrates the fact that worki ng c reatively is as much about knowing what to do but also i ncludes responding to the i mage and being open to trying new approaches and ideas. Use the Curves laye r to d arken the i mage even more by fi rst pulling the h ighlight and then the midpoint down, as seen i n Figu re 10-40. W h ile still in Curves, use the pull-down menu to select the blue channel and increase the blue component, as seen in Figure 1 0-4 1 .
Figure 10-38. The "dayfor night" effect enhance., thefeeling ofsadness at the child�' grave.
"'"
Remove Distractions I
379
Photo Filler 1
Curves 1
Figure 10-39. S.
380
T ransfe r the layer mask from the blue Photo Filter layer to the Curves layer by pressing Option/Alt as you drag the layer mask from the blue layer to the Curves layer. Click Yes to the "Replace Layer Mask?" pop-up w indow to create the work-in-progress image seen i n Figure lO-42.
6.
With the topmost layer active, use Cmd-Option-Shiftl Ctrl-Alt-Sh i ft + E to merge all the worki ng layers up onto a new layer.
7.
Select Filter-> B l u r->Gaussian Blur and use an aggressive setting; in this example we used 1 5. Change the layer blending mode to Darker (CS2 or before) or Darker Color if using CS3, as seen in Figure 1 0-43.
I Chapter 1 0 : Enhancing Focus
Figure 10-40. The Curves layer darkens the image to create a more somber -mood.
8.
Add a layer mask to the blurred layer and coarsely paint over the grave of the child. To increase the transition of blurred to sharp use a soft-edged brush, followed by an aggressive dose of the Gaussian B l u r filter t o create t h e image seen i n Figure 10-44.
9.
To add the final pol ish to the image, select the enamel picture of the little boy with the Elliptical Marquee tool, and use Eclit->Copy Merged to copy all v isible layers, fol lowed by Edit->Paste, which d rops the l ittle boy exactly on top of the original i mage.
Figure 10-42. Transferring the layer 112as/( insures pelfect registration a nd edge comistency.
Laver 1
Curves 1
Figure
10-41.
Increasing the blue component in the scene.
10. Select Filter->Other->High Pass with a setting of 3. Change the layer blending mode to Overlay, Soft Light, or Hard Light to sha rpen the enamel photograph. In this example, we used Overlay at 1 00 percent, as seen i n Figure 10-45, to give him a hint of emphasis. All the blu r and focus enhancement methods a re best used in support of your i nitial vision, and we've learned that not every image needs a lot of heavy-handed mani pulation when a touch of subtle enhancement is much more effective.
Figure 10-43. Softening the merged layer in combination with the Darl(er Color blending mode conveys a sense ofthe mother's view ofher child's grave.
Remove Distractions I
381
S h arpen with F i n esse
Not sharpening an image enough is just a s bad a s over sharpening an i mage and, as discussed in Chapter 4, we practice a th ree-step sharpening process of i n pu t sharpening, creative sharpening, and output sharpening. T he creative sharpening includes the blurring and softening techn i ques d iscussed up to this point. Creative sharpening is used to d raw the v iewer's eye to the subject of your photograph, and it is always applied selectively. I mage sharpeni ng algorithms fin d an i m age edge and increase the contrast along that edge by exaggerating the light/dark relationshi p of the edge. Over-sharpened i mages have an unpleasan t halo-like look that has given d igital imaging a bad reputation. In fact, over-sharpening an i m age is probably worse than not sharpe n i ng an i mage at all. To minimize the possibi l i ty of undesired sharpening artifacts: •
Figure 10-44. Blocking the softeningfrom affecting the gravefocuses your eye on it.
Work on a duplicate layer and use layer masks to conceal the sharpening and prevent it from a ffecting undesired areas. Control the sharpeni ng effect with l ayer opacity, and use the Luminosity blendi ng mode to avoid exaggerating color fri nging.
•
•
•
•
•
•
Avoid sharpe n i ng shadow areas, as they do not contain a lot of edge i n formation and often do contain noise, which is something you rarely want to sharpen. Use the Blend If sliders to sharpen the m i d tones without affecting the shadows o r h ighlights, as described in Chapter 6 in the H igh Pass Contrast Enhancement section. Print your sharpening tests with the printe r and paper that you'll use for the final document to see the actual sharpening effect. Avoid resizing or retouching a sharpened file. When you send files to a service bureau, tell them i f you've al ready applied sharpening o r you want them to do it. Too much sharpening can be just as bad as no sharpening. I n Photoshop CS3, choose F i l ter-Convert for Smart Filters to non-destructively experi ment, adjust, and refine the filter settings.
Photoshop comes with a variety of built-in sharpening filters that range from the most basic and rarely used to very sophisticated with tremendous control as listed in this overview that i ncludes in-depth i n formation on the U nsharp Mask and Smart Sharpen filters. Figure 10-45. Thefinal touch is to shalpen the enamel picture of the little boy.
382
I Chapter
1 0:
Enhanci ng Focus
Filter-->Sharpen o
o
o
o
Sharpen: A brusque filter that offers no control-we never use it. Sharpen More: Sharpens more than Sharpen, which you cannot control, which is why we never use it.
Inside the S m art Sharpen Filter o
o
Sharpen Edges: We never use this one either, as the interface offers no control. Sharpens only edges while preserving the overall smoothness of the image. Unsharp Mask: The venerable sharpening filter, which (unlike its name) does an excellent sharpening job of sharpening by allowing control over the Amount (strength), Radius (width), and Th reshold (details) that are sharpened. Behind the scenes, first the picture is smoothed or blurred slightly (unsharpened) and the contrast is reduced to conceal fine noise. Then the unsharp mask filter i n creases the contrast of adjacent pixels to create the sharpening effect. o
o
o
•
o
Amount: Low Amount settings introduce low levels of contrast to neighboring pixels with high Amount settings adding h igher contrast levels.
Smart Sharpen: A fantastic sharpening filter that is supported by Smart Filters. It can be slow to apply, but the control offered is well worth the wait. The Smart Sharpen filter enables you to set which sha rpening algorithm is used and to control the amount of sharpening i n the shadow and h ighlight a reas. Plus, by clicking the small floppy d isk-like button, you can save successful settings for reuse, which will speed up your i mage workflow.
Remove: Sets the sharpening algorithm used to sharpen the image from the following three options.
•
Radius: The most crucial of the three controls, Radius determines how many pixels on either side of an edge will be exaggerated. The strength of the Radius setting is influenced by the resolution of the image, with high-resolution images tolerating higher Radius settings. Set the radius large enough to create distinct borders, but not so large as to create a noticeable outline around objects.
Threshold: Determines the tonal value d i fferences needed to recognize an edge. For example, if you set the threshold to 6, neighboring pixels with gray values of 1 20 and 1 24 will not be sharpened because the d i fference between them (4) is below the threshold. However, neighboring pixels with values of 1 20 and 130 will be sharpened, because the difference of 1 0 is above the threshold. We usually set our Threshold values between 3 and 7 to avoid sharpening noise, skin pores, and subtle shadows and delicate highlight transitions.
Rad i us: Deter m i nes the number of pixels surrounding the edge pixels a ffected by the sharpening. The greater the Radius value, the wider the edge e ffects and the more obvious the sharpening-meaning the more visible the l ight/dark halo will be, which is something you want to avoid.
o
• o
Amount: Sets the amount of sharpening. A h igher value increases the contrast between edge pixels, giving the appearance of greater sharpness.
Gaussian Blur is the method used by the Unsharp Mask filter and is the fastest. Lens Blur detects the edges and detail in an image, and provides finer sharpening of detail and reduced sharpening halos. In our opinion, Lens Blur produces the best results. Motion Blur attempts to reduce the effects of blur d ue to camera or subject movement. If you plan on using the Motion Blur option, measure the angle of blur with the Measure tool before entering the Smart Sharpen interface, and enter the measured angle in the Angle control.
More Accurate: Processes the file twice for a better removal of blurri ng. Do not use this option on grainy o r noisy files as i t has a tendency to over-crunch unwanted image textu re.
Smart Sharpen Advanced Mode This mode uses all of the previously addressed basic mode settings and enables you to reduce dark or l ight sharpening halos. o
o
o
Fade Amount: A d j u sts the amount of sharpe n i ng in the h ighlights or shadows-lOO-percent fade conceals the sharpening, and decreasing the fade reveals the sharpening amount. Tonal Width: Controls the range o f tones i n the shadows or h ighlights that are modified . Lower values restrict the adj ustments to the darker areas for shadows and only the l ighter regions for highlight correction. Think of this as the spread-how far out do you wan t the sharpening to take place ? Lower settings are more delicate. Rad i us: Controls the size of the area around each pixel that is used to determ ine whether a pixel is in the shadows or the highlights. Moving the slider to the left specifies a smaller a rea, and moving i t to the right specifies a larger a rea.
Sharpen with Fi nesse I
383
Filter-Other The true challenge in working with the Smart Sharpen filter is to find the right balance between the Amount and the Radius settings. For digital camera files or scans with minimal grain, use high
In addition to the commonly used sha rpen ing filters, the Custom and High Pass filters can also be used to sharpen and bl ur i mages. •
Amount settings between 75 and 200 and low Radius settings between 0.3 and 1 .5 . This may look too crisp onscreen, but you have to print the file to see the results that really matter. For images with noticeable grain, use low Amount settings of 1 0 to 40 and high Radius settings of 1 0 to 20 to help avoid exacerbating noise or grain.
•
Custom: Please see the Custom Filter sidebar later in this chapter. High Pass: I s an edge detection fil ter that works on duplicate o r merged layers. Use a setting between 1 and 3, change the layer blending mode to Overlay, Soft Light, or Hard Light, and adjust layer opacity to create the desi red e ffect.
Figure 10-46. Sharpening the textural details emphasizes the plant's interej'ting biology.
384
I Chapter
1 0:
Enhancing Focus
" AJ'
Sharpening the Details Selective sharpeni ng d raws the v iewer's eye in and should be subtle without being d istracting. In t he example seen in Figure 10-46, Katrin photographed the lily with an extremely shallow depth of field and, by selectively sha rpeni ng the pistil, was able to enhance the original photograph to match her original photographic i n tention. In this example, Katrin started with a Smart Object from a camera raw file; to fol low along, you'll start by converting the background layer to a Smart Filter layer, something which is not necessary when starting with a Smart Object from a camera raw file.
1.
Select Filter-Convert for Smart Filters.
2.
Select Filter-Sharpen-Smart Sharpen, and since this file was captured with a medium-format d igital camera back and has very l ittle noise, use a high Amount with a low Radius setting, as seen in Figure 1 0-47.
Figure 10-48. Only the areas ofthe image where the layer mas/( is white will be shClipened. 5.
To i nclude more of the flower in the sharpeni ng pass, use a white brush to paint i n additional sharpeni ng. To exclude a reas from being affected, use a black brush on the mask.
6.
To avoid an artificial "this a rea is sharper than this one" look, use F i l te r-Blu r-Gaussian Blur with a setting of 3 to 5 on the layer mask to blur the transition between what has been sharpened and what has not been sha rpened, as viewed in Figure 1 0-49, which shows the Smart Filter mask.
Figul'e 10-47. The high Amount and low Radius values sharpen the detail without creating aggravating !loise. 3.
To sharpen only the pistil, select the pistil of the flower with the Quick Selection tool or the Lasso tool. There is no need to be absolutely perfect as it is easier to refine the mask with the Brush tool in the following step than to spend a lot of time making a perfect selection.
4.
Choose Select-I nverse, and, with the Smart Sha rpen mask active, select Edit-Fi l l with Black, and then Select-Deselect to create the i mage seen i n Figure 1 0-48.
Figure 10-49. OptionIAlt-clicl( the layer masl( icon to view the layer masl(.
Adding Sparkl e and S h een In portrait retouching, maintaining the sparkle in the subject's eyes keeps the portrait l i vely. In the example seen in Figure 1 0-50, Katrin softened the skin and sharpened the eyes with separate H igh Pass layers as described here.
Sharpen with Fi nesse I
385
Figure 10-50. Shmpening the eyes creates the illusion ofmoisture and sparkle.
(' C"ml'lfl" , 1 m .I',,,d(
--'El ch10_po rtrait.jpg
386
1.
Duplicate the background layer, select Fil ter-> Other->High Pass, and use a setting o f 5 , Change the layer blending mode to Overlay, and select I mage-> Adj ustments->Desaturate to remove u n wanted color from this layer. Name this layer "sharp."
2.
Duplicate this sharp layer, change the blending mode to Soft Light, and select I mage-> Adjustments->Invert to create a subtle and beautiful softening e ffect. Name this layer "soft," and turn its v isibility off.
3.
Click the sharp layer and Option/Alt-click the layer mask button to add a black layer mask. Use a 50-percent hardness white brush on the layer mask to paint ove r the eyes, which will add the sparkle of the sharp layer.
4.
To transfer an exact duplicate of the sharp layer mask while i nverting it, press Cmd Option-ShiftlCtrl-Alt-Shift, and d rag it into the layer slot of the soft layer, as seen i n Figure 1 0-5 1 .
I Chapter
1 0:
Enhancing Focus
Figure 10-51. Each layer mas/( hides and reveals the desired sharpening or blurring effect.
Experiment with adding sparkle to i mages with water d roplets or high tech equipment; kicking up the h ighlights with sharpening can make your images pop off the page.
Softer Can Be Sharper As odd as it sounds, you can use the Gaussian Blur filter to sharpen an i mage. I learned the basics of the following technique from Dan Margulis, who is an expert at color correction, offset printing, and working with the Lab image mode. I n a nutshell, the follow i ng technique uses the Gaussian Blur fi lter on dupl icate fi les and layers with a variety of blending modes to create fi nely detailed edge masks, which control the darkening and l ighteni ng effect to make the i mage sharper, as seen in Figure 10-52 .
JtI ch10_ostridge.jpg 1.
Open an i mage, select I m age--+Duplicate, and name the duplicate fi le Dark Edges.
2.
Dupl icate the background layer of the duplicated file twice.
3.
Select I m age--+Dupl icate, and name the d u plicate fi le
Light Edges. 4.
On the Light Edges i m age, click the middle layer and Filter--+Gaussian B l u r with a setting of ] .5, and change the layer blending mode to Darken.
Figure 10-52. Shalpenillg without using any sha,peningfilters.
Sharpen with Finesse I
387
5.
Click the topmost layer, change the blend ing mode to Difference, and flatten the i mage. T t should be very dark with a few delicate white outli nes, as seen in Figure 1 0-53.
6.
Duplicate the background layer once, change the blending mode to Screen, and duplicate this layer again to strengthen the edge lines, as seen in F igu re 1 0-54. Flatten this file and save it.
7.
Retu rn to the Da rk Edges file and, on the middle layer, use the Gaussian Blur fi l te r with a setting of 2 and change the layer blend i ng mode to Lighten.
8.
Click the topmost layer and cha nge the layer blending mode to Difference. Flatten the i mage.
9.
Invert this image with Cmd/Ctrl + T and dupl icate the background layer once, change the blend i ng mode to Multiply, and dupl icate this layer agai n . Flatten the file, and save i t to create the fil e seen i n Figure 1 0-55 .
,�
, ;\V.', , '
' .,
Figure ]0-53. The initial Lighten Edgesfile will show thefaintest of out/illes.
Figure 10-54. Duplicating the layel:; and using the Saeen blend mode strengthens the white lines.
388
I Chapter 1 0 : Enhancing Focus
On the Dark Edges file, choose Select---> A ll fol lowed by Edit---> Copy. Click the original file and choose Edi t---> Paste, and then change the layer blending mode to Multi ply.
11. On the Light Edges file, choose Select---> A ll followed by Edit---> C opy. Click the original file and choose Ed it---> P aste, and then change the layer blending mode to Screen to create the i mage seen in Figure 1 0-56.
The Custom Filter The primitive grid i nterface h ides a powerful tool used to create darkening, l ightening, bl urring, and sha rpening e ffects. The cente r box describes the brightness of the pixel bein g a nalyzed, with lower n umbers bei n g d a rker and higher numbers lighter. I nserting n u m bers i nto the adjacent boxes calculates the pixels agai nst each other while simulta neously creating i n terestin g H igh Pass, emboss, blur, and sharpen ing effects. [ a n Albert posted both the math and very useful tips (as featured here) on how to use the custom filter: http://ia n-albel't.comlgmphicsl customjilters.php. We've posted Custom kernels you can load to sta rt experimenting with blurring and sharpening effects. General guidelines for using the custom filter: •
•
•
•
Adjust the opacity of each layer to create a pleasi ng contrast and sharpening of the file.
Creati ng two fi les to increase selective contrast may seem l ike a lot of effort, but after running through the steps a few times they will become faster and easier to apply. In the end, the results of this fi ne-edged contrast defi nition technique are ideal for very detailed i mages. Soften then sharpen with the Custom Filter [ n the traditional dark room, if a c h rome was too contrasty to print, a photographer would make an i n termediary negative by sandwiching the ch rome with black-and-white fi l m ; between t h e t w o pieces of film w a s a piece of tissue paper or glassine that would soften the resulting black-and-white fi lm i mage. After processing the black-a nd-white film, the photographer would sandwich the softer black-and-white film with the contrasty c h rome to make a print with a reduced but reproducible dynamic range that also looked sharper than the original c h rome.
I f the sum of the weights d i v ided by the scale is I , then the offset should be 0. I f the sum of the weights is 0, then the offset should be 1 2 8. [f the sum of the weights d iv ided by the scale is I , then the offset should be 2 5 5 . The i mage will be inverted from what it would be if the sum were + 1 .
Figure 1 0-56. Adju>'ting the opacity ofeach layer control>, the final >,harpenil'lg effect. 12.
The scale should be equal to the sum of all the weights.
•
•
The sum of the weights d i v ided by the scale should generally not be more than + I or less than - I . Otherwise, the i mage will come out very dark or very bright. The scale can't be less than 1 (you can't d i vide by zero). If the fi l tered i mage is very dark, the scal e is probably too high. If it's very light, the scale is probably too low.
High-end glamour retoucher C h ris Tarantino taught us that blurring an i mage ever so slightly and then sha rpening with the custom fi lte r creates an a rtifact-free sharpening effect that he uses for high-end cosmetic and celebrity retouching. The example seen i n Figure 1 0-57 is perfectly i n focus, yet the eyes and hat could be sha rper to enhance the crispness of the winter's day.
Sharpen with Fi nesse I
389
FIgure 10-57. Blurring before shclJpening helps avoid UllS1ghtly shalpening artifacts.
JEJ ch10_snow-boy.jpg
Custom OK
JEJ ch10_C ustomBlurSharpen
fIl II 11 11 II ( Cancel II fII f2l fIl II ( load .. . II f2l f8l f2l 11 ( Sav.. .. II fII f2l fIl II � Pr.vi.w 11 11 11 11 11 Scale: � Offset: II
To work along with the Custom Filter exercise, either download the Ch 1 0_Custom BlurSha rpen fi l te r presets and extract them to your hard drive or use the numeric values shown in the ill ustrations to create similar effects. 1.
2.
390
Select Filter�Convert for Smart Filters. If work ing with earlier versions of P h otos hop, please dupl icate the background layer. Select Filter�Other�C ustom and load Blur50%.AFS, o r use the setti ngs seen in Figure 1 0-58 to gently soften the image. Concentrate on an area with fine detail such as eyelashes. Click OK.
I Chapter 1 0 : Enhancing Focus
El
100"
l!1
Figure 10-58. Use the custom filter to soften the image before shCllpening.
3.
Select Filter--->Other--->Custom, and load Sharpen200%. AFS, or use the settings seen in Figure 1 0-59. To j udge the effects, turn off both Smart Filters by clicking the view icon next to the Smart Filte r layer mask. To v iew the effect of a single filter, click the respective view Icon. Custom
00 III II II II II ( Canc.1 ) 11 1-31 1-61 1-31 11 ( load... ) 11 1-61 [SBl 1-6I 11 ( Sav 11 1-31 1-61 1-31 11 � P,.v;.w 11 11 11 11 11 Scal.: [221 Offs.t: II •...
EI
100%
I±I
Figure 1 0-59. Sharpening after blurring is useful to avoid unsightly shClljJening artifacts aLongfine, detailed edges. 4.
Use a large, soft-edged black brush on the Smart Filter layer mask to paint over the little boy's skin so that it remains soft, as seen i n Figure 1 0-60. If working with earlier versions of Photoshop, add a layer mask and paint over the little boy's skin to protect it from the sharpening.
�
W
The beauty of the Smart Filters is that you can a d j ust or load new settings to i m p rove the final outcome, which allows for tremendous creative freedom to experiment and improve your Images.
Focusing on H istory The h istory feature keeps track of your steps as you work, and in combination with the H istory brush, enables you to paint back and forth between d i fferent versions of your images. J ust k nowing that the H istory palette lets you step backwards is very reass u ri ng as you delve i n to creative experimentation. To take advantage of History, i n General Preferences--->Performance increase the n u m be r of History states to at least 50- 1 00. I n the H istory palette, use the fly out menu to make sure that in the History Options "Allow Non-Linear H i story" is selected . In the example seen i n Figure 1 0-6 1 , Katrin softened the entire i mage, and then used the History brush to paint the original flowers back in as follows.
'1J ch10_ B r i de.jpg 1.
Duplicate t h e background layer, select Filter--->Other---> High Pass, a n d use a very high setting. I n this example we used 50.
2.
Change the layer blen d i ng mode to Soft Light and Cmd/Ctrl + I to invert the sharpening to c reate a beautiful soften i ng effect. Click the "Create a new snapshot" button (it looks like a l ittle camera i n the H i story palette), a n d name i t sofi, as shown i n Figure 1 0-62 .
3.
Flatten the i m age and click the small square next to the original s napshot i n the H istory palette. Use a soft edged H istory brush to paint over the flower bouquet to bring it back i nto focus, as seen in Figure 1 0-63 . I n case you paint back parts o f h e r d ress, click the History source next to the soft snapshot and pai n t the softer areas back.
Smart Filters
!!lJ Custom � Custom
Figure 10-60. lvlas/(ing tI,e shclIjJenillgfrom the boy�' s/(In ma/(es the eyes look even shclIpel:
Sharpen with Fi nesse I
391
Figure 10-61. Softening the dress emphasizes the bouquet. , I{"rk {;,.I,'Noise--->Add Noise with a high setting of 30 and Monochrome selected. To make the noise clumpier (that is, less d igital-looki ng), she chose Filter--->Noise--->Median and with a setting of I , as seen in Figure 1 0-72 .
3.
To apply only the noise to the i mage background, she took advantage of an existing layer mask to create a mask for the crucifi x that blocked out the noise from impacting the delicate relief on the crucifix (Figure 1 0-73) .
Take a few minutes t o dood le o n the neutral gray Noise layer to create your own look and feel. In addition to the Median filter, you can add a nother pass of Noise or use the Motion Blur filter to create a streaked effect. Try using the Unsharp Mask filter to make the noise even c runchier. Si nce you're working on a separate layer, your i magi nation is the only limit to the textural grain effects you can create.
Yo u r Vis i o n C o m e s Fi rst
Relish experi mentation, which always includes making so-called mistakes, which i n reality are simply wonderful learning opportunities. Be open, doodle, and push the digital tools and techniques fu rther than we can imagine. You r craft, your image, and your vision come fi rst.
Figure 1 0- 73. }3locl(ing the noisefrom affecting the delicate relief emphasized the crucifix appropriately.
YOU l' Vision Comes F i rst I
399
I n d ex
Sym b o l s
B
16-bit editing, 133
Background layer, 146
for initial i n frared
9X Media, 1 8
backgrounds
conversion, 329
black-and-white infrared, 328-332
black-and-white a d j ustment
a d j usti ng depth of field, 368-3 7 1
custo m izing contrast with
A
bl u r r i ng, 373-375 blurring and d a rkening, 375-379
glowing h ighl ights, 330
additive color, 230
darkening, 364-368
Add Noise filter, 327, 395
gra i n layer, 33 1
depth of field effects, 3 7 1 -3 73
adjustment layers, 134
Photoshop i n fra red vs. real
i m proving, 362-382
blending modes, 1 48- 1 5 1 color tinting, 1 68 dodging and burning with e m pty adj ustment layers and blending modes, 1 79- 1 83 layer masks, 1 76- 1 78 controlling visibility with black, white, and gray, 1 76 feathering mask edges with Gaussian B l u r, 1 77 I nvert com m a n d , 1 77 Levels and C urves, 8 Photo Filter, 2 5 8-259 reticulation, 337 Adobe Camera Raw, 8 Adobe Gamma utility, 2 8 Adobe Photoshop Lightroom, 8 Adobe RGB ( 1 998 ) , 29 Advanced Controls, 35 aperture theory, 54 Append File Extension, 23 Apple 30-inch Cinema Display, 1 8 Apple RGB, 29 architectural perspective, 1 1 1 Ask Before Saving Layered TIFF Files, 24 atmosphere, 1 2 Automatically Launch Bridge, 2 1
remov i ng un wanted elements, 362-364 tinti ng, d a rkeni ng, and blurring, 3 79-3 82 back ups, 1 9 barrel distortion, 109 batch renaming, 56 Beckelman, Mark, 298 Bicubic Sharper, 2 1 Bicubic Smoother, 2 1 black-and-white conversion, 1 5 1-159
Camera Raw, 1 53-1 56 saving grayscale settings with your i mage fi l es , 1 54 saving presets, 1 55 G rayscale Mixer, 1 53 P hotoshop, 1 56- 1 59' Black and W hite a d j ustment layer, 1 56 Chan nel Mixer, 1 58
cu rves, 33 1
i n frared, 332 black-and-white photographs
accenti ng with color, 1 66-1 7 1 blending o riginal i mage colors w i th black and white, 1 67 color tinti ng, 1 68 Black and White adjustment dialog, 328 black and white point, 8 Black Point compensation, 35 Blacks slider, 73 Blend If sliders, 1 93-1 97 blending modes, 1 46- 1 5 1 , 1 92-194
adj ustment l ayers, 1 48-1 5 1 darkening and buil d i ng density with Multi ply, 1 48 i n c reasing contrast with Overlay o r Soft Light, 1 50 l ighte n i ng with Screen, 1 49 Color Dodge mode, 1 47 d a rkeni ng, 1 46
Photoshop Lightroom, 1 5 1 - 1 53
Darken m ode, 1 46
raw files, 1 5 1 - 1 56
Darker Color m ode, 1 47
Ta rgeted Adj ustment Tool, 1 52
dodging a n d burning with em pty
black-and-white film photography
adj ustment l ayers and
Zone System, 1 30
blending modes, 1 79- 1 83
blending modes (continued)
em pty adj ustment layers and
c
I m age A d j ustments tabs, 68-87 Basic, 69-72
contrast-enhancing blending
Cache Levels, 26
modes, 1 92
calibration, 28
enhancing contrast, 1 48 fad ing i m ages, 342 Hard Ligh t mode, 1 4 8 l ighte n i ng, 1 47 Lighten mode, 1 47 Lighter Color mode, 1 47 Linear Dodge mode, 1 47 modify i ng contrast, 1 95 Multiply mode, 1 47, 1 48 Overlay mode, 1 48, 1 50 overv iew, 1 46- 1 48 Screen mode, 1 47, 1 49 Soft Light mode, 1 48, 1 50 targeted adj ustments with blending modes, 1 93
(see also mode blendi ng)
Camera Raw, 20, 62-68
adj usting white balance, 69 Auto Adj ustments how they i m pact previews in B ridge, 68 Auto ad j ustments, 66 black-and-white conversion, 1 53- 1 56 saving grayscale settings with your i mage files, 1 54 saving presets, 1 55 Blacks sl ider, 73 Brightness slider, 74 Camera Calibration tab, 84 c h romatic aberration, 83 Clarity control, 74
blown-out highlights, 53
clipping warni ng, 65
blurring, 12
Color Noise Reduction slider,
fad i ng i mages, 342 Bridge, 20
a utomatically launching, 2 1 Camera Raw (see Camera Raw)
8 1 , 93 color temperature, 304 con trast strategies for adj usting, 78
i n i tial adj ustments to raw files, 59
Contrast slider, 74
keywords, 57
Crop tool , 63
label and star system, 57
cu rves, 78
brightness, 230
(see also l u mi n osity) Brightness slider, 74 Brightness slider (Levels) , 138 budgeting, 1 6 burning, 1 1 , 1 7 4
controll i ng, 78 saving, 79 Darks slider, 76 Defringe command, 83 Exposure controls, 70 Fill Light slider, 72 h ighlight recovery, 7 1 H ighl ights slider, 76 h istogram and settings menus, 66 HSLlG rayscale tab, 8 1 -82 I m age Adj ustments, 66 Basic, 66 Camera Calibration, 66 Detail, 66 HSLlGrasycale, 66
Camera Calibration, 84 HSLlGrayscale, 8 1 -82 Lens Correction, 82 P resets panel, 84 Split Ton i ng, 82 Tone C urve, 76-79 i n pu t sharpening, 1 07 Detail slider, 1 08 Maski ng, 1 08 i nter face, 62-66 Lens Correction tab, 82 Lights slider, 76 Lumi nance Smoothing slider, 80, 93 noise reduction, 80 opening JPG files in, 23 Parametric cu rve, 76 Point Curve, 77 Presets panel, 84 Preview, 63 Recovery sl ider, 7 1 Saturation slider, 76 setting Exposu relWh i te Point with bright h ighl ight, 72 setting Exposure/White Point w i thout bright highl ight, 73 Shadows s l ider, 76 sharpening, 80 split toning, 1 63 Split Ton i ng tab, 82 Straighten tool, 63 Tem perature slider, 69 Tint slider, 69 Tone Curve tab, 7 1 , 76-79 tools, preview and cli pping displays, 63 Vibrance control, 76 vignetti ng, 83 W h i te Bala nce tool, 69 Work flow Options, 66 camera raw files, 8
Levels, 236-240 identifying the brightest and da rkest points, 238 mid tone color casts, 242-246 neutralizing highlight and shadow points, 239 prepa ring workspace, 237 setting eyed ropper target val ues, 237 local, 259-277 changing color, 263-278
color noise, 8
conceal i ng, 9 5 Color Noise Reduction slider, 8 1 , 93 Color Picker, 2 1 Color Range, 269-270 color reference targets, 54 color settings, 27-29
sav ing, 36 Color Settings Comment dialog box, 36 Color Settings dialog box, 29
Colo r Management Policies, 33-35
blending (see mode blendi ng)
Cu rves, 260-263
Convert to Wor k i ng RGB, 33
com ponen ts, 229-232
Match Color com mand (see
O ff, 33
converting to black-and-white (.;ee black-and-white conversion) enha ncing, 1 0 hue (see h ue)
Match Colo r command) Replace Color command, 267-268 Replace Color command with
lum inance (see l u m i na nce)
Color Ra nge, 269-270
saturation (see saturation)
m a n ua l l y neutralizing m i dtone
selective i m p rovement, 1 2 subjectiv i ty of, 229 what color is, 288 work i ng with, I I Color blending mode, 262 color casts
color fil l layers to neutralize color casts, 262-263 correcting with no neutral guide, 244 manually neutralizing mid tone color cast, 243
color cast, 243
Preserve Embedded Profiles, 33 P rofi l e M i smatches and Missing P rofiles, 34 color tags, 28 color temperature, 301-3 1 3
c reative raw processing, 304 adj usti ng w hite balance, 306
neutral color, 232-234
fal se color temperature, 304
setting neutral highl ights with
m ulti ples of raw, 3 1 0-3 1 3
Cu rves, 24 1 target neutrals, 234 with neut rals, 235 Color Dodge mode, 147 color enhancements, 25 1-259
H ue/Satu ration command (see H ue/Saturation com mand) color fringing, 9 color ingredients, 230-232
not-neutral w h i te balance, 308 v ariations on i mage, 309 neutral , 303 color tints, 168, 291
adj ustment layers and layer m asks, 1 68 com b i n i ng, 296-298 painting on em pty layer, 1 70 Column Size, 27
Color Channels in Color, 22
additive color, 230
complementary colors, 23 1
color correction
complementary colors, 23 1
component colors, 232
big picture, 232
com ponent colors, 232
contact sheets, 5
color e n ha ncements (see color
subtractive color, 23 1
enhancements) correcting with no neutral guide, 244 Cu rves (see Cu rves, global color correction; local color correction)
color management, 27-29
mon itor calibration and profi l i ng, 28 profiles, 28 th ree key pri nciples to a profiles fu nctional ity, 28
existing neutrals, 235
work i ng spaces, 28
global , 232-246
(see also Col or Settings d ialog box,
Lab color (see Lab col or)
Color Ma nagement Policies) ColorMatch RGB, 30
continuous tone devices, 44
compa red to i n k jet pri nters, 45 sca nning resolution, 47 contone devices, 44 contrast, 1 1 , 1 3 1-132
controll i n g i n C u rves (j"ee Curves, control l i n g contrast) enhancing, 1 0 w i th blending modes, 1 48 i n c reasing with Overlay or Soft Light, 1 50 stra tegies fo r adj usting, 78
(see also contrast and exposu re control)
channels/contrast
I ndex I
403
Find Dark and Light Colors
contrast and exposure control
(Auto Color), 250
blending modes (see blending
target colors and clippi ng, 25 1
modes) fil l fl ash tec h n i q ues, 203 H igh Pass con trast e n hancement, 1 96- 1 98 l um i nance masks, 203-205
gray point eyed ropper, 242 Lab color (see Lab col o r, Curves) local color correction, 260-263 alteri ng color, 263-265 color fill layers to neutral i ze color
digital sensors, 53 digital wallets, 56 digital zone system, 130
comparable to Levels, 1 4 1
Darken mode, 146
contro l l i ng contrast, 1 43-1 46
Darker Color mode, 147
direct digital printers, 44
darkroom
Display Calibrator Assistant, 28
precise control with l ockdown points, 1 44 reverse S-curve, 1 43 S-curve, 1 43 finding tones and working with control poi nts, 1 42 global color correction automatic correction, 249-2 5 1 Enhance Monochromatic Contrast ( Auto Contrast), 250 Enhance Pe r Channel Contrast (Auto Levels), 250 example I: Clea n i ng the pond, 246-248 exam pie 2 : Backyard witch, 248-250
404
Surface B l u r fi l ter, 323-324
resolution, 49-50
casts, 262-263
command)
D i ffuse Glow fi l ter, 324-325 G aussian B l ur filter, 322-323 digital cameras, 20, 49-62
h istogram, 1 40
Shadows/Highlights command
diffusion and soft focus, 322-325
I I n dex
d igital (see d igital darkroom) traditional (see traditional dark room) darkroom special effects, 332-337
creative accidents, 332 solarization, 332-335
Distort control, 9 distortion
correcting, 9 distractions, removing, 360-382
essential cleanup, 360-362 i mproving backgrounds, 362-382
Darks slider, 76
adj usting depth of field, 368-3 7 1
Defringe command, 83
blurri n g and darkening, 3 75-379
Delkin, 1 9
b l u rring backgrounds, 373-375
depth o f field
darkening, 364-368
adj usti ng, 368-3 7 1
depth of field effects, 3 7 1 -373
effects, 37 1 -373
removing unwanted elements,
dFine, 101 Diffuse Glow filter, 324-325, 330
362 tinti ng, darkening, and blurri ng, 379-382
contrast and exposure control/d istractions, I-emoving
distressed effects, 339-348
faded black-and-white photos, 34 1 -343 fading to dark with blur layers and blend modes, 342-343 faded color photos, 339 matching a modern i m age to vintage print, 343
editing, non-destructive, 133-134, 1 74-224
1 6-bit edi ting, 1 33
file duplication, 6
a d j ustment layers, 1 34
file formats
dodging and burning (see dodging and burni ng)
Smart Filters (>"ee Smart Filters)
focus, 345
starting with raw fi le, 1 33
DNG Converter, 62 DNG files, 60
advantages, 60 conversion, 56, 60-62
black-and -white i n frared (>"ee c reative edge (see creative edge e ffects)
Photoshop Lightroom, 6 1
dark room special effects (>"ee
(>u aL>"o dodging and burni ng) dodging and burning
empty adj ustment layers and blending modes, 1 79- 1 83 automating with a n action, 1 8 1 - 1 83 DoubleSight, 1 8 downloaded images, multiple copies, 56 downsampling, 2 1 duplication o f images, 6 dustbusting images, 1 0 dust spotting, 1 1 8 DxO Optics Pro, 8 dynamic range, 208-2 1 4
combining m ulti ple exposures with layer masks, 2 1 2-2 1 5 multi ple raw processing o f a single exposu re, 208-2 1 2
darkroom special e ffects) d i ffusion and soft focus (>"ee d i ffusion and soft focus) distressed (>"ee d i stressed e ffects) fil m grain (see film gra i n ) fi l ters Vee fi l ters) reticulation (see reticulation) sepiaesque d i ffusion ton i ng, 325-327
labels (see labels), 56 rati ngs (see ratings) raw fi les (see raw files, i nitial a d j ustments) file preparation, 8-10, 93-127
c h romatic aberration (>"ee c h romatic aberration ) c ropping (see cropping) noise red uction (see noise red uction) optical and perspective d i stortion
(see optical and perspective distortion) spotting and cleanup (see spotting
layer masks (see layer masks) neutral fi ll layer, 1 74- 1 75
color and contrast, 344 texture, 345
F
and clean u p) fill flash techniques, 203
Fill Light control, 2 03 Fill Light control, 203 Fill Light slider, 72 film grain, 327-328
A d d Noise fi l ter, 327 creating softer grain to m i m i c a n e n l a rged negative, 3 2 8 film recorders, 44 film scanners, 1 9 filters
A d d Noise, 327 cool i ng, 259 Custom (see Custom fi l ter) D i ffuse Glow, 324-325, 330
keywords, 57
I ndex I
405
filters (continued)
Gaussian B l u r (see Gaussian B l u r fil ter) H igh Pass (see H igh Pass fil te r) Lens B l u r (see Lens B l u r filter) sha rpen i ng (see sharpening, filters) Smart B l u r (see Smart B l u r filter) Sola ri ze, 332 Surface B l u r, 323-324 Texturizer, 338 warmi ng, 259 flare, 1 2 1
removing, 1 25 flatbed scanners, 1 9
H
histogram (Curves ) , 140 histograms, 1 1 , 52-54
Hand tool, 36-38
History & Cache, 25
Hard Light mode, 148
History feature, 391-394
hardware considerations
History Log, 2 1
back ups, 1 9
hue, 230, 288-301
bus speed, 1 7
color tints, 29 1
CPU speed, 1 7 d igital cameras, 20 hard d rives, 1 7 monitors, 1 8 cali b ration, 1 8
focus, 1 2
mouse and mouse pad, 1 9 p ressu re-sensitive tablet, 1 9
feathering m ask edges with, 1 77 global corrections, 134- 1 5 1
Cu rves (see Cu rves) Levels (see Levels) global enhancements, 1 0 Gorman, Greg, 1 6 0 gradation masking, 1 1 Gradient tool, 1 22, 177, 207, 296, 300, 30 1 , 3 1 6, 370, 373 gray, significance of, 232 gray point eyedropper
neutralizing i m age, 242 grayscale images, 134 Grayscale Mixer, 1 53, 163 Gray working spaces, 32 Greene, Jeff, 366 Gretag Macbeth, 28 Gretag Macbeth Eye One Display 2, 1 8 guide print, 2 8 Guides, Grids, and Measure tool, 1 14
e n hancing natural colors, 295 neutral density and color bIens, 298-3 1 4 warming i mages, 293-294 selectively, 293 Hue/Saturation, 10, 1 1
adj usting i n d i v idual colors, 254-25 5 adj ustment l ayer monochromatic toning, 1 59
accessing Merge to HDR, 2 1 6
Gamma slider (Levels ) , 1 38
selectively, 294
RAM, 1 7 HDR
G
combining, 296-298 cooling i m ages, 294-295
second, 1 8
Free Transform command, 9
Full Screen Mode, 36
th ree areas o f concern, 53
hard drives, 17
Equal i ze H istogram, 2 1 9
enhancing color with, 2 5 1-2 5 7 fine-tun i ng spec i fi c color adj ustments, 2 55-260
Exposure and Gamma, 2 1 8 H ighlight Compression, 2 1 9 Local Adaptation, 2 1 9 adj usting Local Adaptation curve, 220
Lightness slider, 253 creative uses for, 253 Saturation slider, 252 vibrance versus saturation, 257-258
combining m ul tiple bracketed exposures with Merge to HDR, 2 1 5-224 combining m ultiple H D R i m ages, 223 Merge to HDR pre-fl ight check, 2 1 6 photographic considerations for H D R photography, 2 1 5 Healing Brush tool, 360, 364, 379 healthy work habits, 17 high-bit files, 50
K keystoning, 109 keywords, 56, 57-59 Kodak Professional Durst Lambda printer, 45
basics, 2 78 Curves, 279-282 building custom correction fi l ter, 283 center point, 279 controll i ng brightness and contrast with the L channel, 280 remov i ng a color cast with the a and b chan nelsRemoving a color cast with the a and b cha nnels, 2 8 1 when t o use, 284 labels, 56-58
object a n d observation, 228-229 Lighten mode, 147
d istortion, 1 1 1
auto correction options, 1 39
L
light
correction, Levels) comparable to Curves, 1 4 1 eyed roppers, 1 39 gray poi n t eyed ropper, 242 I nput Levels, 1 3 5-1 38 B rightness slider, 1 38 c l i pping d isplay, 1 37 Gamma slider, 1 3 8 I n put h ighlight slider, 1 36 I nput shadow slider, 1 35 Midtone slider, 1 38 setting the black point, 1 35 setti ng the white point, 1 36 Output Levels, 1 38- 1 39 Levels and Curves adjustment layers, 8
tonal m a s k i ng, 1 04-1 05 Mass Engineered Design, 1 8 Match Color command, 270-279
fine-tu n i ng with Cu rves, 276 I mage Options, 2 72 with selections, 2 73 , 2 75 Maximized Screen Mode, 36 Maximize PSD and PSB File Compatibility, 24 Memory Usage, 25 menu bar, hiding, 36 Microsoft Expression Media
color and l ight, 3 1 4-3 1 5 l ay e r m asks, 3 1 5-3 1 9 Monaco Optix XR, 1 8 monitors, 1 8
calib ration, 1 8, 2 8 second, 1 8
I ndex I
407
monochromatic toning, 1 5 9-162
G reg Gorman's warm-tone process, 1 60-1 62 H u e/Saturation adjustment layer, 1 59
o
Move Grid tool, 1 1 0
offset printing, 46
determ ining scan n i ng resolution, 46 scan n i ng resol ution, 47 optical and perspective distortion, 109-1 1 7
Multiply mode, 147, 148, 364, 376, 379, 380, 388
black-and-white conversion,
observational exercise, 4
More Options button, 35 mouse and mouse pad, 1 9
Photoshop CS3, 20
architectural perspective, I I I ba r rel distortion, 1 09
N
chromatic aberration, 1 09
naming files, copies, 6
keystoning, 1 09
NEC Spectra View Reference 2 1 , 1 8
Lens Correction filter (see Lens
i m age level ness, 1 09
needs, identifying, 1 6 .NEF format, 60 neutral color, 232-234
existing neutrals color correction, 235 natural and target, 233 target neutrals color correction, 234 neutral density and color blends, 298-3 1 4 neutral fill layer, 174-175 Nikon 5000 series scanner, 20 Nikon 9000 series scanner, 20 Nikon Capture, 8 Nikon Coolscan 4000, 47 Nik Software, 1 0 1 noise, 93
avoid i ng before pressing shutter, 1 05 valuable, 394-399
(see also color noise; l u m i na nce noise), 95 Noise Ninja, 1 0 1 noise reduction, 8, 93-106
in Camera Raw, 80 masking (see maski ng) stronger, 96 third-party, 1 0 1 noise reduction solutions, 53 Noiseware, 1 0 1
Correction fil ter) lens distortion, I I I perspective cropping, 1 1 5 pi ncushion d istortion, 1 09 vignetti ng, 1 09 optical distortion, correcting, 9 original quality, 43 over-exposed highlights, 93 Overlay mode, 148, 150
1 56-1 59 B lack and W h i te adjustment layer, 1 56 Channel Mixer, 1 58 Camera Raw (see Camera Raw) Color setti ngs (see color settings) Curves, 1 40 [ nterface prefe rences Append File Extension, 23 Ask Before Saving Layered T I FF Files, 24 Color Channels i n Color, 22 Enable La rge Document Format (.psb), 24 Ignore EXIF sRGB Tag, 23 I mage Previews, 23 Maximize PSD and PSB File Compatibil ity, 24 P re fe r Adobe Camera Raw o r J PE G fi les, 23 Show Menu Colors, 22 Show Tool Tips, 22 Performance preferences
(see also optical and perspective distortion) Perspective Crop tool, 9 PhaseOne Capture One, 8 Photo Filter, 1 1
adj ustment layers, 258-285 photographer, being a better, 4
P a i n t i n g Cursors, 2 6 Scratch Disks, 2 5 Plug-ins settings, 2 7 Preferences, 20-22 Automatical l y Launch B ridge, 2 1 Color Picker, 2 1 Export Cl i pboa rd, 2 1 H istory Log, 2 1 I mage I nterpolation, 2 1 Show C lipping option, 1 4 1 spotting, 1 1 9 Tra nsparency & Gamut preferences Gamut Warning, 27 Tra nsparency Settings, 26 Units & Rulers settings Col u m n Size, 2 7 U n its, 27
408
! I ndex
monoch romatic ton i ng/Photoshop ( 5 3
Photoshop Lightroom, 20
black-and-white conversion, 1 5 1 - 1 53 color temperatu re, 304 DNG conversion in, 6 1 i nitial adj ustments to raw files, 59 keywords, 57 label and star system, 57
ProPhoto RGB
advantages for raw files, 30 d isadvantages, 3 1 PSB files, 24
compatibil i ty with PSD fi les, 24 PSD files
com patibility with PSB files, 24 purchasing better equipment than you currently need, 1 6
split toni ng, 1 63 PictureCode, 1 0 1 plug-ins, 2 0 , 2 7 Point Curve, 77
quality
Prefer Adobe Camera Raw or JPEG files, 23 pressure-sensitive tablet, 19 printers, 20, 43-47
batch (see batch renaming) Replace Color command, 1 1 , 267-268 researching equipment, 1 6 resolution
pri nter resolution versus i mage resolution, 44
printer resolution versus i m age resol ution, 44
(j-ee a/so continuous tone devices) printer technology, 43 printing digital files, 43 print size, 43 prioritizing equipment purchases, 1 6 processing digital images, 6 professional digital labs, 45 profiles (color management) , 28
mIssIng Assign Profile option, 35 Assign Working RGB option, 35 Leave As I s (Don't Color Manage) option, 35 three key principles to a profiles functionali ty, 28 warni ngs, 34 Convert Document's Colors to the Working Space option, 34 Discard the Embedded Profile (Don't Color Manage) option, 35 profile mismatch, 34 Use the Embedded P rofile option, 34
Photoshop Lightroom/scan ners
reticulation, 335-337
R
with adj ustment layers, 337
RAID (Redundant Array of Independent Disks), 1 7 RAM, 1 7 ratings, 7, 56-58 RAW Developer, 102 raw files, 50
black-and-white conversion, 1 5 1 - 1 56 color temperatu re, 304
A dobe R G B ( 1 998), 29 A pple RGB, 29 ColorMatch RGB, 30 G ray working spaces, 32 ProPhoto R G B advantages for r a w files, 3 0
false, 304
d isadvantages, 3 1
multiples of raw, 3 1 0-3 1 3 not-neutral white balance, 308 variations on i mage, 309
sRGB, 30 Ruler tool, 1 1 5
white balance, 306 exposu re considerations, 53 h igh-bit data, 5 1 i n itial adj ustments, 56, 59 new formats, 60 noise reduction, 93 spotti ng, 1 1 8 underexposed, 54 raw processors, 8
s Saba tier, Armand, 332 Sabatier effect, 332 saturation, 230
versus vibrance, 258 Saturation slider, 76 scanners, 1 9
recomposing images, 1 1 4
I ndex I
409
scanning, 8
before, 43 calculating optimal scan size for i n k jet printers, 44 determining resolution for offset printing, 46
in Camera Raw, 80
Spot Healing Brush, 1 1 9
input o r capture sharpening, 1 06
spotting and cleanup, 1 1 7-126
input sharpening, 1 06
c h romatic aberration, 1 23
input sharpening with Adobe
d ust spotting, 1 1 8
Camera Raw, 1 07 m i n i m i zing unwanted effects, 382
editorial cleanup, 1 20 fla re, 1 2 1
fil m original, 42
output or print sharpeni ng, 1 06
Photoshop, 1 1 9
once for d i fferent purposes, 47
selecti ve, 385
spotting in raw files, 1 1 8
resolution s u m m a ry, 46
soften then sharpen, 389
Scratch Disks, 25
uneven, 9
Screen mode, 147, 149
Sharpen More filter, 383
screen modes, 36
Show Menu Colors, 22
seeing, the art of, 4
Show Tool Tips, 22
Selective Color, 1 0
Skew control, 9
Selective Color adjustment layer, 1 1
Smart Blur filter, 375
selective enhancements, 1 1
Smart Filter layers, 323
sepiaesque diffusion toning, 325-327
Smart Filters, 95, 199
service providers
Smart Sharpen filter, 383
send i ng images to, 44 shadow point, 1 1 shadows, 93 ShadowslHighlights command, 199-200
Amount, 200 Black Clip, W hite Clip, 202 Color Correction, 2 0 1 i n practice, 202 M idtone Contrast, 2 0 1
Advanced Mode, 383 Fade Amount, 383 Radius, 383 Tonal Width, 383 Amou nt, 383 More Accurate, 383 Radius, 383 Remove, 383 soft focus (see diffusion and soft focus)
creative or interpretive sharpeni ng, 1 06 filters Custom , 384, 389 H igh Pass, 384 Sharpen, 383
consumables, plug-ins, and education, 20 solarization, 332-335
custom solarization with Cu rves, 334 solarization curves with color photos, 334 Solarize fi l ter, 332 Solarize filter, 332 sparkle and sheen, 385-387
soften then sharpen, 389 split toning, 162-166
Sharpen Edges, 383
Camera Raw, 1 63
Sharpen More, 383
h ighl ight and shadow masks, 1 63
Smart Sha rpen, 383
i rregular toning by painting in
Unsharp Mask, 383
v ignetting, 1 2 1 spotting images, 1 0 Spot Working Space, 3 3 sRGB color profile in EXIF metadata, 23 Standard Screen Mode, 36 straightening and correcting perspective, 1 1 4 straightening images, 9 Straighten tool, 1 1 0 Straighten tool (Camera Raw ) , 63 subtractive color, 23 1 Surface Blur filter, 96, 323-324
color correction, 234 temperature (see color temperature) Temperature slider, 69 textures, 1 2, 337-339
blending i n to skies, 348-349 classic, 337 combining photos with art paper, 347 creating for use in Photoshop, 346-347 custom textures with Texturizer fil ter, 338 watercolor paper, 338 Texturizer filter, 338 thermal printers, 44
versus saturation, 258 Vibrance control,76 viewfinder, 49
tonal information,42
vignetting,83,109,117
tonality,10,93
visual components, identifying,183
tonal masking,104-105
visual story, enhancing, 356
tone,130-131
capturing good tonal detail, 53 enhancing, 10
Wacom's Cintiq 2lUX,19 Warming Filter (81),293
shadows (see shadows)
Warming Filter (LBA),296
photographs with color (see black-and-white photographs, accenting with color), 166-171 monochromatic toning, 159-162 split toning, 162-166 traditional darkroom
Zone System,130 zone system,digital,130 Zoom tool,37,38,261
w
highlights (see highlights)
accenting black-and-white
Zenview,18
zoom ratios,37
midtones (see midtones) toning and split toning effects,159
z
warming filters, 259 warming images, 293-294 Warp control, 9 watercolor paper texture,338 white balance
adjusting in Camera Raw, 69 color temperature, 306 White Balance tool,69,233-236,304, 306,308 workflow,digital darkroom,7 working environment,16-17 working spaces,28
CMYK,32 Gray,32
u
RGB (see RGB working spaces)
U.S. Web Coated (SWOP)v2,32 underexposed images,92 uneven sharpness,9 Unsharp Mask filter,383, 397,399 unwanted artifacts,93 upsamp1ing,2l Use Dither for 8-bit images,35
