IFIP Advances in Information and Communication Technology
345
Editor-in-Chief A. Joe Turner, Seneca, SC, USA
Editorial Board Foundations of Computer Science Mike Hinchey, Lero, Limerick, Ireland Software: Theory and Practice Bertrand Meyer, ETH Zurich, Switzerland Education Arthur Tatnall, Victoria University, Melbourne, Australia Information Technology Applications Ronald Waxman, EDA Standards Consulting, Beachwood, OH, USA Communication Systems Guy Leduc, Université de Liège, Belgium System Modeling and Optimization Jacques Henry, Université de Bordeaux, France Information Systems Jan Pries-Heje, Roskilde University, Denmark Relationship between Computers and Society Jackie Phahlamohlaka, CSIR, Pretoria, South Africa Computer Systems Technology Paolo Prinetto, Politecnico di Torino, Italy Security and Privacy Protection in Information Processing Systems Kai Rannenberg, Goethe University Frankfurt, Germany Artificial Intelligence Tharam Dillon, Curtin University, Bentley, Australia Human-Computer Interaction Annelise Mark Pejtersen, Center of Cognitive Systems Engineering, Denmark Entertainment Computing Ryohei Nakatsu, National University of Singapore
IFIP – The International Federation for Information Processing IFIP was founded in 1960 under the auspices of UNESCO, following the First World Computer Congress held in Paris the previous year. An umbrella organization for societies working in information processing, IFIP’s aim is two-fold: to support information processing within its member countries and to encourage technology transfer to developing nations. As its mission statement clearly states, IFIP’s mission is to be the leading, truly international, apolitical organization which encourages and assists in the development, exploitation and application of information technology for the benefit of all people. IFIP is a non-profitmaking organization, run almost solely by 2500 volunteers. It operates through a number of technical committees, which organize events and publications. IFIP’s events range from an international congress to local seminars, but the most important are: • The IFIP World Computer Congress, held every second year; • Open conferences; • Working conferences. The flagship event is the IFIP World Computer Congress, at which both invited and contributed papers are presented. Contributed papers are rigorously refereed and the rejection rate is high. As with the Congress, participation in the open conferences is open to all and papers may be invited or submitted. Again, submitted papers are stringently refereed. The working conferences are structured differently. They are usually run by a working group and attendance is small and by invitation only. Their purpose is to create an atmosphere conducive to innovation and development. Refereeing is less rigorous and papers are subjected to extensive group discussion. Publications arising from IFIP events vary. The papers presented at the IFIP World Computer Congress and at open conferences are published as conference proceedings, while the results of the working conferences are often published as collections of selected and edited papers. Any national society whose primary activity is in information may apply to become a full member of IFIP, although full membership is restricted to one society per country. Full members are entitled to vote at the annual General Assembly, National societies preferring a less committed involvement may apply for associate or corresponding membership. Associate members enjoy the same benefits as full members, but without voting rights. Corresponding members are not represented in IFIP bodies. Affiliated membership is open to non-national societies, and individual and honorary membership schemes are also offered.
Daoliang Li Yande Liu Yingyi Chen (Eds.)
Computer and Computing Technologies in Agriculture IV 4th IFIP TC 12 Conference, CCTA 2010 Nanchang, China, October 22-25, 2010 Selected Papers, Part II
13
Volume Editors Daoliang Li Yingyi Chen China Agricultural University EU-China Center for Information & Communication Technologies (CICTA) 17 Tsinghua East Road, Beijing, 100083, P.R. China E-mail: {dliangl, chenyingyi}@cau.edu.cn Yande Liu East China Jiaotong University College of Mechanical and Electronic Engineering Shuanggang Road, Nanchang, 330013 Jiangxi, China E-mail:
[email protected] ISSN 1868-4238 e-ISSN 1868-422X e-ISBN 978-3-642-18336-2 ISBN 978-3-642-18335-5 DOI 10.1007/978-3-642-18336-2 Springer Heidelberg Dordrecht London New York Library of Congress Control Number: 2010942867 CR Subject Classification (1998): I.2.11, H.4, C.3, C.2, D.2, K.4.4
© IFIP International Federation for Information Processing 2011 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting: Camera-ready by author, data conversion by Scientific Publishing Services, Chennai, India Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Preface
I want to express my sincere thanks to all authors who submitted research papers to the 4th IFIP International Conference on Computer and Computing Technologies in Agriculture and the 4th Symposium on Development of Rural Information (CCTA 2010) that were held in Nanchang, China, 22–25 October 2010. This conference was hosted by CICTA (EU-China Centre for Information & Communication Technologies, China Agricultural University); China Agricultural University; China Society of Agricultural Engineering, China; International Federation for Information Processing (TC12); Beijing Society for Information Technology in Agriculture, China. It was organized by East China Jiaotong University. CICTA focuses on research and development of advanced and practical technologies applied in agriculture and aims at promoting international communication and cooperation. Sustainable agriculture is currently the focus of the whole world, and the application of information technology in agriculture has become more and more important. ‘Informatized agriculture’ has been the goal of many countries recently in order to scientifically manage agriculture to achieve low costs and high income. The topics of CCTA 2010 covered a wide range of interesting theories and applications of information technology in agriculture, including simulation models and decision-support systems for agricultural production, agricultural product quality testing, traceability and e-commerce technology, the application of information and communication technology in agriculture, and universal information service technology and service systems development in rural areas. We selected 352 best papers among those submitted to CCTA 2010 for these proceedings. It is always exciting to have experts, professionals and scholars getting together with creative contributions and sharing inspiring ideas which will hopefully lead to great developments in these technologies. Finally, I would like also to express my sincere thanks to all the authors, speakers, session chairs and attendees for their active participation and support of this conference.
October 2010
Daoliang Li
Conference Organization
Organizer East China Jiaotong University
Organizing Committee Chair Yande Liu
Academic Committee Chair Daoliang Li
Conference Secretariat Lingling Gao
Sponsors China Agricultural University China Society of Agricultural Engineering, China International Federation for Information Processing, Austria Beijing Society for Information Technology in Agriculture, China National Natural Science Foundation of China
Table of Contents – Part II
Food Safety and Technological Implications of Food Traceability Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hailiang Zhang, Xudong Sun, and Yande Liu
1
Function Design of Township Enterprise Online Approval System . . . . . . Peng Lu, Gang Lu, and Chao Ding
11
Application of GPS on Power System Operation . . . . . . . . . . . . . . . . . . . . . Chunmei Pei, Huiling Guo, Xiuqing Yang, Bin He, Wei Liu, and Xuemei Li
18
Greenhouse Temperature Monitoring System Based on Labview . . . . . . . . Zhihong Zheng, Kai Zhang, and Chengliang Liu
23
Image-Driven Panel Design via Feature-Preserving Mesh Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Baojun Li, Xiuping Liu, Yanqi Liu, Ping Hu, Mingzeng Liu, and Changsheng Wang
30
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber on the Cooling Rate during Vacuum Cooling . . . . . . . . Tingxiang Jin, Gailian Li, and Chunxia Hu
41
Inspection of Lettuce Water Stress Based on Multi-sensor Information Fusion Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hongyan Gao, Hanping Mao, and Xiaodong Zhang
53
Measurement of Chili Pepper Plants Size Based on Mathematical Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yun Gao, Xiaoyu Li, Kun Qi, and Hong Chen
61
Methodology Comparison for Effective LAI Retrieving Based on Digital Hemispherical Photograph in Rice Canopy . . . . . . . . . . . . . . . . . . . . . . . . . . Lianqing Zhou, Guiying Pan, and Zhou Shi
71
Molecular Methods of Studying Microbial Diversity in Soil Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liu Zhao, Zhihong Ma, Yunxia Luan, Anxiang Lu, Jihua Wang, and Ligang Pan Monitoring the Plant Density of Cotton with Remotely Sensed Data . . . Junhua Bai, Jing Li, and Shaokun Li
83
90
VIII
Table of Contents – Part II
Motion Blurring Direction Identification Based on Second-Order Difference Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Junxiong Zhang, Fen He, and Wei Li
102
Multi-agent Quality of Bee Products Traceability Model Based on Roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yue E, YePing Zhu, and YongSheng Cao
110
NIR Spectroscopy Identification of Persimmon Varieties Based on PCA-SVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shujuan Zhang, Dengfei Jie, and Haihong Zhang
118
One Method for Batch DHI Data Import into SQL-Server: A Batch Data Import Technique for DateSet Based on .NET . . . . . . . . . . . . . . . . . . Liang Shi and Wenxing Bao
124
Optimal Sizing Design for Hybrid Renewable Energy Systems in Rural Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yu Fu, Jianhua Yang, and Tingting Zuo
131
Overall Layout Design of Iron and Steel Plants Based on SLP Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ermin Zhou, Kelou Chen, and Yanrong Zhang
139
Performance Forecasting of Piston Element in Motorcycle Engine Based on BP Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rong Dai
148
Performance Monitoring System for Precision Planter Based on MSP430-CT171 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lianming Xia, Xiangyou Wang, Duanyang Geng, and Qingfeng Zhang
158
Pervasive Agricultural Environment Monitoring System Based on Embedded Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hu Zhao, Sangen Wang, and Dake Wu
166
Precipitation Resource Potential in Mountainous Areas in Hebei Province Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zheng Liu, Yanxia Zheng, and Zhiyong Zhao
177
Precision Drip Irrigation on Hot Pepper in Arid Northwest China Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Huiying Yang, Haijun Liu, Yan Li, Guanhua Huang, and Fengxin Wang Study on Thermal Conductivities Prediction for Apple Fruit Juice by Using Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Min Zhang, Zhenhua Che, Jiahua Lu, Huizhong Zhao, Jianhua Chen, Zhiyou Zhong, and Le Yang
185
198
Table of Contents – Part II
IX
Prediction of Agricultural Machinery Total Power Based on PSO-GM(2,1, λ, ρ Model) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Di-yi Chen, Yu-xiao Liu, Xiao-yi Ma, and Yan Long
205
Prediction of Irrigation Security of Reclaimed Water Storage in Winter Based on ANN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jinfeng Deng
211
Progress of China Agricultural Information Technology Research and Applications Based on Registered Agricultural Software Packages . . . . . . Kaimeng Sun
218
Quantification Research on Different Load Weight-Bearing Running Biochemical Indexes of Rats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Huaping Shang
227
Rapid Determination of Ascorbic Acid in Fresh Vegetables and Fruits with Electrochemically Treated Screen-Printed Carbon Electrodes . . . . . . Ling Xiang, Hua Ping, Liu Zhao, Zhihong Ma, and Ligang Pan
234
Regional Drought Monitoring and Analyzing Using MODIS Data—A Case Study in Yunnan Province . . . . . . . . . . . . . . . . . . . . . . . . . . . Guoyin Cai, Mingyi Du, and Yang Liu
243
Regression Analysis and Indoor Air Temperature Model of Greenhouse in Northern Dry and Cold Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ting Zhao and Heru Xue
252
Remote Control System Based on Compressed Image . . . . . . . . . . . . . . . . . Weichuan Liao Analysis of the Poverty-Stricken Rural Areas’ Demand for Rapid Dissemination of Agricultural Information—Taking Wanquan County in Hebei Province as an Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaoxia Shi and Yongchang Wu
259
264
Research and Analysis about System of Digital Agriculture Based on a Network Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Duan Yane
274
Research and Development of Preceding-Evaluation System of Rural Drinking Water Safety Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lian He and Jilin Cheng
283
Research of Evaluation on Cultivated Land Fertility in Xinjiang Desert Oasis Based on GIS Technology—Taking No. 22 State Farm as the Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ling Wang, Xin Lv, and Hailong Liu
290
X
Table of Contents – Part II
Research of Pest Diagnosis System Development Tools Based on Binary Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yun Qiu and Guomin Zhou
300
Research of Soil Moisture Content Forecast Model Based on Genetic Algorithm BP Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Caojun Huang, Lin Li, Souhua Ren, and Zhisheng Zhou
309
Research of the Measurement on Palmitic Acid in Edible Oils by Near-Infrared Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hui Li, Jingzhu Wu, and Cuiling Liu
317
Research on a Heuristic GA-Based Decision Support System for Rice in Heilongjiang Province . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ran Cao, Yushu Yang, and Wei Guo
322
Research on Docking of Supply and Demand of Rural Informationization and “Internet Digital Divide” in Urban and Rural Areas in China . . . . . . Zhongwei Sun, Yang Wang, and Peng Lu
329
Research on Evaluation of Rural Highway Construction in Hebei Province . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Guisheng Rao, Limeng Qi, Runqing Zhang, and Li Deng
339
Research on Farmland Information Collecting and Processing Technology Based on DGPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weidong Zhuang and Chun Wang
345
Research on Fertilizer Efficiency of Continuous Cropping Greenhouse Cucumber Based on DEA Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaohui Yang, Yuxiang Huang, Shuqin Li, and Sheng Huang
351
Design and Implementation of Crop Recommendation Fertilization Decision System Based on WEBGIS at Village Scale . . . . . . . . . . . . . . . . . Hao Zhang, Li Zhang, Yanna Ren, Juan Zhang, Xin Xu, Xinming Ma, and Zhongmin Lu
357
Research on Influenced Factors about Routing Selection Scheme in Agricultural Machinery Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fan Zhang, Guifa Teng, Jie Yao, and Sufen Dong
365
Research on Informationization Talented Person Training Pattern of the Countryside Area in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yang Wang and Zhongwei Sun
374
Research on Quality Index System of Digital Aerial Photography Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wencong Jiang, Yanling Li, Yong Liang, and Yanwei Zeng
381
Table of Contents – Part II
XI
Research on Quality Inspection Method of Digital Aerial Photography Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaojun Wang, Yanling Li, Yong Liang, and Yanwei Zeng
392
On RFID Application in the Tracking and Tracing System of Agricultural Product Logistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weihua Gan, Yuwei Zhu, and Tingting Zhang
400
Research on Rough Set and Decision Tree Method Application in Evaluation of Soil Fertility Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Guifen Chen and Li Ma
408
Research on the Method of Geospatial Information Intelligent Search Based on Search Intention Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jingbo Liu, Jian Wang, and Bingbo Gao
415
Research on the Theory and Methods for Similarity Calculation of Rough Formal Concept in Missing-Value Context . . . . . . . . . . . . . . . . . . . . Wang Kai, Li Shao-Wen, Zhang You-Hua, and Liu Chao
425
Research on Traceability System of Food Safety Based on PDF417 Two-Dimensional Bar Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shipu Xu, Muhua Liu, Jingyin Zhao, Tao Yuan, and Yunsheng Wang
434
Research and Application of Cultivation-Simulation- Optimization Decision Making System for Rapeseed (Brassica napus L.) . . . . . . . . . . . . Hongxin Cao, Chunlei Zhang, Baojun Zhang, Suolao Zhao, Daokuo Ge, Baoqing Wang, Chuanbao Zhu, David B. Hannaway, Dawei Zhu, Juanuan Zhu, Jinying Sun, Yan Liu, Yongxia Liu, and Xiufang Wei Residue Dynamics of Phoxim in Pericarp, Sarcocarp and Kernel of Apple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yunxia Luan, Hua Ping, and Ligang Pan Risk Analysis of Aedes triseriatus in China . . . . . . . . . . . . . . . . . . . . . . . . . . Jingyuan Liu, Xiaoguang Ma, Zhihong Li, Xiaoying Wu, and Nan Sun Risk Assessment of Reclaimed Water Utilization in Basin Based on GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yanxia Zheng, Shaoyuan Feng, Na Jiang, and Qingyi Meng
441
457 465
473
Root Architecture Modeling and Visualization in Wheat . . . . . . . . . . . . . . Liang Tang, Feng Tan, Haiyan Jiang, Xiaojun Lei, Weixing Cao, and Yan Zhu
479
Sensors in Smart Phone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chunmei Pei, Huiling Guo, Xiuqing Yang, Yangqiu Wang, Xiaojing Zhang, and Hairong Ye
491
XII
Table of Contents – Part II
Simulation Analyze the Dice and Shape of the Dicer Based on ADAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yingsa Huang, Jianping Hu, Deyong Yang, Xiuping Shao, and Fa Liu
496
Simulation and Design of Mixing Mechanism in Fertilizer Automated Proportioning Equipment Based on Pro/E and CFD . . . . . . . . . . . . . . . . . Liming Chen and Liming Xu
505
Simulation Study of a Novel Algorithm for Digital Relaying Based on FPGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Renwang He, Dandan Xie, Yuling Zhao, and Yibo Yang
517
Simulation Study of Single Line-to-Ground Faults on Rural Teed Distribution Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wanying Qiu
521
Single Leaf Area Measurement Using Digital Camera Image . . . . . . . . . . . Baisong Chen, Zhuo Fu, Yuchun Pan, Jihua Wang, and Zhixuan Zeng
525
Sliding Monitoring System for Ground Wheel Based on ATMEGA16 for No-Tillage Planter—CT246 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lianming Xia, Xiangyou Wang, Duayang Geng, and Qingfeng Zhang
531
Soil Erosion Features by Land Use and Land Cover in Hilly Agricultural Watersheds in Central Sichuan Province, China . . . . . . . . . . . . . . . . . . . . . . Zhongdong Yin, Changqing Zuo, and Liang Ma
538
Spatial and Temporal Variability of Annual Precipitation during 1958–2007 in Loess Plateau, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rui Guo, Fengmin Li, Wenying He, Sen Yang, and Guojun Sun
551
Spatial Statistical Analysis in Cow Disease Monitoring Based on GIS . . . Lin Li, Yong Yang, Hongbin Wang, Jing Dong, Yujun Zhao, and Jianbin He Study for Organic Soybean Production Information Traceability System Based on Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xi Wang, Chun Wang, Xinzhong Wang, and Weidong Zhuang Study of Agricultural Informatization Standards Framework . . . . . . . . . . . Yunpeng Cui, Shihong Liu, and Pengju He
561
567 573
On Countermeasures of Promoting Agricultural Products’ E–Commerce in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weihua Gan, Tingting Zhang, and Yuwei Zhu
579
Study on Approaches of Land Suitability Evaluation for Crop Production Using GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Linyi Li, Jingyin Zhao, and Tao Yuan
587
Table of Contents – Part II
Tracking of Human Arm Based on MEMS Sensors . . . . . . . . . . . . . . . . . . . Yuxiang Zhang, Liuyi Ma, Tongda Zhang, and Fuhou Xu Study on Integration of Measurement and Control System for Combine Harvester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jin Chen, Yuelan Zheng, Yaoming Li, and Xinhua Wei Study on Jabber Be Applied to Video Diagnosis for Plant Diseases and Insect Pests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wei Zhang, JunFeng Zhang, Feng Yu, JiChun Zhao, and RuPeng Luan Study on Pretreatment Algorithm of Near Infrared Spectroscopy . . . . . . . Xiaoli Wang and Guomin Zhou Study on Rapid Identification Methods of Transgenic Rapeseed Oil Based on Near Infrared Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shiping Zhu, Jing Liang, and Lin Yan Study on Regional Agro-ecological Risk and Pressure Supported by City Expansion Model and SERA Model – A Case Study of Selangor, Malaysia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaoxia Shi, Yaoli Zhang, and Cheng Peng Study on Relationship between Tobacco Canopy Spectra and LAI . . . . . . Hongbo Qiao, Weng Mei, Yafei Yang, Wang Yong, Jishuai Zhang, and Yu Hua Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI in Inner Mongolia Grassland . . . . . . . . . . . . . . . . . . . . . . . . . . Hongbin Zhang, Guixia Yang, Qing Huang, Gang Li, Baorui Chen, and Xiaoping Xin Study on Storage Characteristic of Navel Orange Based on ANN . . . . . . . Junfang Xia and Runwen Hu Study on the Differences of Village-Level Spatial Variability of Agricultural Soil Available K in the Typical Black Soil Regions of Northeast China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weiwei Cui and Jiping Liu Study on the Management System of Farmland Intelligent Irrigation . . . Fanghua Li, Bai Wang, Yan Huang, Yun Teng, and Tijiu Cai Extracting Winter Wheat Planting Area Based on Cropping System with Remote Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xueyan Sui, Xiaodong Zhang, Shaokun Li, Zhenlin Zhu, Bo Ming, and Xiaoqing Sun
XIII
597
607
615
623
633
641 650
658
667
674 682
691
XIV
Table of Contents – Part II
Study on the Rainfall Interpolation Algorithm of Distributed Hydrological Model Based on RS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaoxia Yang, Yong Liang, and Song Jia Study on Vegetable Field Evaluation Index System for Non-Point Source Pollution of Dagu River Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jinheng Zhang, Junqiang Wang, Yongliang Lv, Jianting Liu, Dapeng Li, Zhenxuan Yao, Xi Jiang, and Ying Liu
700
706
Study on Water Resources Optimal Allocation of Irrigation District and Irrigation Decision Support System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liang Zhang, Daoxi Li, and Xiaoyu An
716
Study on Web-Based Cotton Fertilization Recommendation and Information Management Decision Support System . . . . . . . . . . . . . . . . . . . Yv-mei Dang and Xin Lv
726
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
735
Food Safety and Technological Implications of Food Traceability Systems Hailiang Zhang, Xudong Sun, and Yande Liu East China Jiaotong University, School of Mechatronics Engineering, 330013, China
Abstract. Food safety has become an important food quality attribute.Both food industry and authorities need to be able to trace back and to authenticate food products and raw materials used for food production to comply with legislation and to meet the food safety and food quality requirements. Traceability is increasingly becoming a necessary task in the food industry which is mainly driven by recent food crises and the consequent demands for transparency in the food chain. This is leading to the development of traceability concepts and technologies adapted to different food industry needs. The content of this paper include several aspects such as overseas food traceability system present conditions and development, food traceability system present conditions, problems and prospect in China, put forward the main measures of pushing on food traceability system of china. Keywords: food traceability, quality, safety, technology.
1 Introduction The demand for food traceability has significantly expanded in the last few years all over the world with increasing incidence of food-related safety hazards and scares such as footh-and-mouth disease, mad cow disease, microbial contamination of fresh produce, dioxin in poultry which greatly decline consumer confidence on food safety. There can be found several definitions for traceability, such as “the ability to follow the movement of a food through specified stages of production, processing and distribution”(Codex Alimentarius,2004), “the ability to trace the history, application or location of that which is under consideration” or “when considering a product, traceability can be related to the origin of materials and parts, the processing history, the distribution and location of the product after delivery”(International Standardization Organization (ISO)). The EU Regulation 178/2002 describes it as “the ability to trace and follow a food, feed, food-producing animal or substance intended to be, or expected to be incorporated into a food or feed, through all stages of production, processing and distribution”.(E.Abad,2009) The term “food traceability” can be traced back to 1986, the year that the first case of mad-cow disease (Bovine Spongiform Encephalopathy, BSE) was reported in the UK. Four years later, the government of UK started a committee to survey the cause and origins of BSE, using the traceability system of cattle production, which is the embryonic form of the current food traceability system. Food traceability has placed responsibilities on producers, processors, D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 1–10, 2011. © IFIP International Federation for Information Processing 2011
2
H. Zhang, X. Sun, and Y. Liu
caterers and other handlers in the supply chain to ensure food safety because foods are mostly composed of more than one ingredient and have often been composed of a series of processes. Raw material producers, ingredients/packaging suppliers, distributors, storage operators, retailers, points of sale, shops, and transporters are all relative partners of the food traceability system. Within the food industry, traceability implies the ability to trace and follow feed, food, and food producing through all stages of production, processing, and distribution. As a result, making a complete traceability system will need enormous resources and effort. Traceability is essential, particularly with raw materials, to establish that control procedures have been applied and are effective. There are good examples of where traceability has had a specific approach, e.g. beef labelling and genetically modified materials.(M.F Stringer,2007) Traceability systems can be considered as a bridge between producers and consumers, since the details of where the products come from and how they are marketed is available for those who are concerned.
2 Technological Implications of Food Traceability Systems Food traceability system is highly knowledge-intensive and increasingly informationdriven. Technological innovations are necessary to reduce transaction costs and facilitate the production of top quality,safe and traceable products to meet consumer demands. Technological innovations are needed for product identification,process and environmental characterization, information capture, analysis, storage and transmission. These technologies include hardware (such as measuring equipment, identification tags and labels) and software (computer programmes and information systems). 2.1 Food Product Identification Technology A major character of the food traceability system is the ability to trace-back the history and the physical location of the food products. To achieve these, accurate labeling is essential. The simplest technology to achieve this is to attach a tag to the surface of the food package and to transfer that data on the tag to the bar code of the food product. (Fig.1)The use of computers and other information technologies have spurred the development of electronic identification (EID) systems, which include electronic tags with chips and scanners for reading, storing and transmitting the data to PCs for analysis and long-term storage. An important attribute of tags is that the materials must be resistant to rough handling and bad weather. Advancements in material science have led the development of tags that are resistant to tear which can withstand harsh environmental conditions.Innovations in geospatial science and technology such as radio frequency technology and mobile tracking devices have the potential for collecting and transmitting data from tags to distant locations for storage and analysis. 2.2 Quality and Safety Measurement Technology The success of traceability is to meet the expectations of the consumer and other stakeholders, the ability to ascertain the location of the food product for effective recall in the event of food quality or safety breach. This requires accurate information
Food Safety and Technological Implications of Food Traceability Systems ODEOH˖ERG\RIFRZH[PDSOH
3
ODEOH˖DIWHUILUVWSURGXFHSURFHVV
Holland Ajax D.A. weight˖ 350g
Holland Ajax D.A. weight˖ 350g
*7,1UHIHUHQFHFRGH˖(- SODFHRIELUWK˖EHOJLXPSODFHRIIHHG˖*HUPDQ SODFHRIVODXJKWHU˖KROODQG VODXJKHUFRGH˖8'+
*7,1UHIHUHQFHFRGH˖(- SODFHRIELUWK˖EHOJLXPSODFHRIIHHG˖*HUPDQ SODFHRIVODXJKWHU˖KROODQG VODXJKHUFRGH˖8'+ SODFHRIILUVWWUDQVSRUW$PHULFDQ
code bar code bar
Fig. 1. Bar code of the food product
on the maturity and quality attributes and safety status of the product, which must be measured and analyzed using appropriate instrument and procedures. Product features such as size, firmness, soluble solids, acidity, flavour, etc, are some of the physical, mechanical and chemical properties that may require measurement. Nondestructive tests based on force sensing, infrared and magnetic resonance imaging can also be used to measure firmness and other internal quality attributes (Linus U. Opara,2003). 2.3 Genetic Analysis Technology The need to preserve the identity of food product and the demand for genetic traceability have led to the development of procedures and measurement devices for the analysis of the genetic constitutions and contamination of foods and other biological products(Giese, J.H. 2001). 2.4 Environmental Monitoring Technology Environmental conditions such as data of temperature and relative humidity collecting, atmospheric composition of the air including pollutants and so on which impact on the quality stability and safety of food products. Instrumented environmental recording devices for monitoring these parameters are available (Linus U. Opara,2003) .environmental monitoring process is shown in Fig.2.
Fig. 2. Environmental monitoring process
4
H. Zhang, X. Sun, and Y. Liu
2.5 Evelopments in Geospatial Science and Technology The integration of geographic information systems (GIS), remote sensing (RS) and global positioning systems (GPS) offers considerable opportunity for the derivation of data related to the food products. These technologies enable data to be remotely collected on a farm, which can be processed, transmitted and presented as product attributes. With respect to food traceability, a vital feature of these technologies is the possibility to map the geospatial variability of selected attributes such as yield, product quality (Bossler, J.D. 2001). 2.6 Web and Database Technology for Food Traceability System Web and database technology relies on the application of appropriate computer system, and which links the food traceability to a central database at the company, national or international level.(Fig.3) Food traceability systems require lots of data uploading to be saved as digital files. The growth in the use of personal computers, with immense processing power, and the continued development of the Internet provides an appropriate environment.The increasing speed and capability of the required communication hardware together with falling prices also contribute to the viability of such a system. Modern personal computers provide a simple means of connecting to the outside world, using software and hardware which are provided with the machine.The system makes use of Internet technology to implement a worldwide solution.It is the physical connectivity of the Internet, as well as the communication protocol (TCP/IP) that is used. The food traceability system can be quickly deployed anywhere in the world. This can be communicated directly to the database server.This process is automated and transparent to the user. After being verified by officially qualified institutions, agricultural food products are labeled with a traceable combination of numbers, just like IDs for foods. Purchasers who purchase goods with these labels can trust the products are officially guaranteed to be of safe and high quality. Terminals set at retail markets are the communication port of producers and consumers. Once the traceable label is scanned on the machine, the detailed traceable information will be showed on the terminal screen. center database of food traceability system
Internet produce process data
base of food produce
sale process data
identity data
data administator
Fig. 3. Center database structure of food traceability system
Food Safety and Technological Implications of Food Traceability Systems
5
2.7 RFID Technology Tags, which contain an integrated circuit chip and antenna, are integrated into objects such that these objects can be identified and their label information can be read. Radio frequency identification involves detecting and identifying a tagged object via radio waves transmitting data from tags to a reader (Jung Lyu Jr.,2009). One of the greatest challenges of implementing food traceability system under certain circumstance lies in the fact that products distribution has a global covering, thus it becomes difficult to precisely trace goods movement throughout the distribution chain. There is a solution that could eliminate these difficulties – an automate gathering of data named radiofrequency identification RFID). Moreover, the RFID tag can store much more information than the linear bar code, and the information can be updated. Sometimes, solutions can be thought in order to combine bar code technology, RFID and vocal recognition, to create a flexible infrastructure which would optimally use the advantages of each technology. (Goodrum, McLaren, & Durfee, 2006; Kwon & Choi, 2008). RFID technology allows the storing of information about all the products that have circulated in a certain container. This type of traceability is very useful for retailers, who can easily locate where to find a certain product, for a rapid delivery. With the help of the RFID tags, a supplier of fresh products (fruit and vegetables, for example) can trace where the goods have been delivered in order to accelerate the payment, or a retailer can be sure that the products are on shelves in the order they were stocked.
3 Overseas Food Traceability System Present Conditions and Development As knowledge and economic grow, the people in developed countries are more concerned about quality and safety of food. Food traceability system is considered as a risk management tool for food safety and is widespread in developed countries such as Japan, the U.S., Canada and many countries in the European Union. Food traceability systems can provide clear, correct sources and marketing routes of food products. With an integrated food traceability system, the government can recall the products immediately and limit the possible loss. In December 2003, the United States developed the statutes of tracking food safety,”Farm Security and Rural Investment Act” requires country of origin labeling for many kinds of food, including perishable agricultural commodities, which required all enterprises involved in food transportation, distribution and import recording their trade information for tracking and tracing back. In addition, the United States also plans to include 70 percent of the cattle in the NAIS (National Animal Identity System) project at the end of 2009. The European Union adopted mandatory traceability actions in food industry since 1st January 2005, Regulation No. 178/2002 establishes that food business operators must label adequately food, in order to facilitate the traceability(Official Journal of the European Union,2002). The Marche Region (Italy) project called SiTRA, aiming to provide food chain stakeholders with a Web platform managing traceability for the main regional food products. Consumers can identify the SiTRA traced products by means a Regional Brand (QM Quality guaranteed by Marche) embedded in the
6
H. Zhang, X. Sun, and Y. Liu
product label, reporting the traceability code to access the Web for traceability information of the product. SITRA platform is currently tracing 8 large food chains (fresh milk, bread, pasta, fresh fish, olive oil, wine, pigs and cattle) all over the Region. In June 2002, the Canadian federal government established an ambitious goal that, before 2008, the country would achieve tracing back 80 percent of agricultural products to its source, supporting the "Brand Canada strategy", of which a mandatory identification system for cattle and beef on July 1, 2002 came into operation. Japan was the first country to introduce a food traceability system in Asia. Due to the occurrence of a series of food safety events, the Japanese government raised food traceability system promotion to the list of important administration policies of Prime Minister Koizumi and expected to accomplish 50% implementation of food traceability in 2007 and 100% by the year 2010.Since 2001, the Japanese government has been promoting the development and use of food traceability systems, and the integration of traceability systems with agricultural risk management systems in order to improve food safety amongst food operators such as producers (farmers), retailers, and manufacturers (Nanseki and Yokoyama, 2008). The supply chain networks for the food industry are reacting both to global trends and to the changes brought about by the continued expansion and ever-deeper integration of the European Union. Ferrer and Findlay (2003) hold the opinion that the drive to unity is strong, but the diversity of competitive practices, labour laws, and regulations across Europe different countries and regions is rich. The supply chain winners will be those who can continue to create new opportunities through the implementation of optimised networks and the development of collaborative partnerships across their extended enterprises. (Ingrid Hunt, 2005).
4 Food Traceability System Present Conditions in China 4.1 Present Conditions It is highly necessary for China to establish traceability systems. For concerning about both domestic food safety and international trade, since 2000 China has been adopting lots of measures and programs to introduce, extend, encourage and even mandate traceability system in food supply chain. In legislation, there are a few specific laws or regulations concerning food safety but little referring traceability before 2001. There have been numerous food contamination events and animal diseases like avianinfluenza, foot-and-mouth disease, bad duck eggs and event of milk in recent years, causing extensive panic among the populace and tremendous losses to the farmers of China. Regarding on substantial advantages such as reducing marketing costs, ensuring product integrity, increasing consumer confidence, China is now developing and implementing food traceability programs throughout (Ministry of Commence of China, 2006). Provide visibility of China’s food safety and quality systems to consumers, importers and governments of world, as well as to domestic businesses and consumers. Enhance the user's confidence on the safety of materials, raw materials and products from China. In order to ensure food safety, Beijing had established a specialized food safety traceability system for the 2008 Olympics, in order to monitor food quality from the origin of production to each stage of processing, packaging,
Food Safety and Technological Implications of Food Traceability Systems
7
transportation, distribution till ultimate consumption. The Beijing Administration for Industry and Commerce (BAIC) and Beijing Food Safety Supervision Office (BFSSO) already founded Beijing Food Safety Traceability System (henceforth BFSTS) based on Capital Food Safety Monitoring System (CFSMS) under the network environment of BAIC. BFSTS consists of one first–level platform and four individual sub–systems. The first–level platform is that Beijing Data Centre for food safety traceability, which is responsible for information collection, analysis, evaluation, tracing, early–alarming. The four sub–systems includes fruit and vegetable, animal products, prepackaged food and Olympic food traceability sub–system. One survey in city markets and rural markets performed in 2006 by Ministry of Commence of China showed that 53.7% of city markets, 32% of supermarkets, 80.4% of wholesale markets of agricultural products, 70.7% of retail markets introduced the above initial measures to foster traceability (Ministry of Commence of China, 2006). More than half of frozen food had already been able to be traced back to origin in 2008. In April 2004, the State Food and Drug Administration and other departments chose meat industry as a pilot industry, started meat and meat products traceability institution construction and system implementation (General Administration of Quality Supervision. 2002). The main tasks include: developing suitable technical standards and Management norms, publishing guidelines for implementing traceability system including "Meat products tracking and tracing Guide" and "Fresh product tracking and tracing Guide”. In June 2004, Administration of national barcode management promoting investigated on vegetable products traceability and started an application project on two vegetable production bases located in Shouguang and Luocheng respectively in Shandong province. The project was successful in food quality control in the origin and enforcing standards of market access, product identification and recall. Integrated with electronic auctions and E-commerce, this project established a traceability system for pollution-free vegetables. Shanghai Livestock Bureau legislated to build digital archives for pigs, cattle, sheep, and the residents can now get access to the egg production information through internet. In August 2008, Beijing had already enforced a food traceability system along the full supply chain for the food supplied for Olympic games to secure food quality and safety. According to the law of People’s Republic of China on food products’ safe quality issued in 2006, all agricultural enterprises must set up production recording that should be authentic and be kept for at least 2 years; otherwise, the transgressor will be penalized not more than 5000 RMB. In addition, individual producers are also encouraged to keep recording within their own production. Such actions are considered as the rudiments of food traceability system in China. On June 26 of 2006, Ministry of Agriculture of China (MOA) promulgated the regulation on animal labeling and their feeding documents establishment in farms (No. 67 Act).In this act, the most key points are as follows: animals such as swine, cattle and sheep/goats in the farms must be gained unique identity code around the China, and it shall be labeled with a special tag embodying a unique code before moved from its region of origin. The feeding enterprises must establish their feeding information documents to record inputs mainly used, such as feeds, feed additives and veterinary drugs as treatment. Traceability on food–produced animals will be started in case of any of the following issues: (a) Some labeling are not in accord with livestock and products from
8
H. Zhang, X. Sun, and Y. Liu
themselves; (b) Livestock and their products have been infected or contaminated by some diseases and virus; (c) There is no quarantine certificate authorized by the accredited institution; (d) Some veterinary drugs and other venomous substance have been used, which are forbidden in terms of related regulations; (e) when serious animal health events occurred; and (f) other situations in which traceability should be applied. 4.2 Problems and Measure of Pushing on Food Traceability System of China Promoting food traceability systems among consumers should focus on obtaining recognition from consumers. Being unfamiliar with this new concept in china, the public may consider “traceability” only as a commercial term, which makes no difference whether the products are traceable or not. Since most consumers don’t understand the principle and value of food traceability systems, the Chinese government should endeavor to educate the public on the concept of food safety and possible food contamination routes. Once the public recognize that the food traceability system is a possible solution which can prevent certain problems and ensure their health, they will be interested in traceable food products and be willing to purchase the traceable agricultural products at a higher price. Recent theoretical literatures provide some useful information relevant to analyzing different consumers’ increasing concerns about food quality and safety knowledge and the effect on food choices. Some studies about consumers show knowledge about food safety tends to increase with age, level of education, and experience in food preparation. These research findings are useful to assist in research of Chinese consumers’ perception toward quality and safety of safe products (Wang Feng, Zhang Jian, 2009). The basis of the traceability system is the detailed data recorded by the producers. In the early days of promotion in China, most of the farmers didn’t consider the traceability system as a constructive policy but rather a complicated and inconvenient one. From their viewpoint, the consumers would not care about the traceable records; the policy was considered just a waste of effort and time. Besides, some older farmers are not educated as well as the young; so even the ordinary paper records are huge obstacles for them, not to mention electronic documents. For companies, compliance to legislation is recognized as the major driving force towards introducing a quality supervision and traceability system. Value added to products through increased consumer confidence may be another important reason. The government should start to educate the producers about the advantages of implementing food traceability systems, popularizing the concept of food safety, making them believe that this could be a wonderful resource with full cooperation and could lead to a more profitable career. The endorsement from consumer to food traceability system is the biggest strength of promotion. Food safety and quality, rather than price, is considered the most important factor affecting food product purchasing decisions of Chinese consumers (Zhang, 2002). Once the consumers agree and favor the traceable products, the rise in profits will bring confidence to the producers. Through mutual trust established via food traceability systems, the benefits of both the producers and the consumers can be ensured. Therefore, food traceability systems
Food Safety and Technological Implications of Food Traceability Systems
9
can be considered as an investment which gives farmers profitable and stable business, and safe food for the public, leading to a promising future for sustainable agriculture.
5 Conclusion The emergence of food traceability system is the result of developments in improving food quality and safety management. Farmers, processors and handlers, and food policy experts need to be aware of future developments in this area to assist them in implementing food traceability systems for their enterprises. The methods for data capture, data exchange, data storage and the integration of the food traceable supply chain are essential for the success of food traceability system. The traceability system of food products, including fish, poultry and meat products, means that the information of a product, from producing and processing to marketing, is recorded and can be traced, “from farm to fork”. If all the food products are implanted with traceability, the consumer can not only query detailed information about the food but also secure themselves from impairment by checking suspicious process in the food supply chain through the food traceability system.
References 1. Codex Alimentarius, Codex Alimentarius Commission. FAO/WHO (2004) 2. Abad, E., Palacio, F., Nuin, M.: RFID smart tag for traceability and cold chain monitoring of foods: Demonstration in an intercontinental fresh fish logistic chain. Journal of Food Engineering, 394–399 (2009) 3. Stringer, M.F., Hall, M.N.: A generic model of the integrated food supply chain to aid the investigation of food safety breakdowns. Food Control 18, 755–765 (2007) 4. US Federal Register: Farm Security and Rural Investment Act of 2002, vol. 68(210), October 30 (2003) 5. Official Journal of the European Communities: Regulation (EC) No 178/2002 Of The European Parliament And Of The Council of 28 January 2002, article 18 (2002) 6. RFID Position Statement of Consumer Privacy and Civil Liberties Organizations, Privacy Rights Learing house (November 30, 2003) 7. Bossler, J.D. (ed.): Manual of geospatial science and technology, p. 664. Taylor & Francis Group plc., UK (2001) 8. Opara, L.U.: Traceability in agriculture and food supply chain: A review of basic concepts, technological implications, and future prospects. Food, Agriculture & Environment 1(1), 101–106 (2003) 9. Giese, J.H.: Lab exhibits promote traceability and safety. Food Technology 55(8), 100, 102–104 (2001) 10. Ferrer, J., Findlay, C.: European supply chain management characteristics and challenges. Ascet achieving supply chain excellence through technology (2003), http://www.ascet.com (accessed 11/09/2003) 11. Hunt, I., Wall, B.: Applying the concepts of extended products and extended enterprises to support the activities of dynamic supply networks in the agri-food industry. Journal of Food Engineering 70(2005), 393–402 (2005)
10
H. Zhang, X. Sun, and Y. Liu
12. Lyu Jr., J., Chang, S.-Y., Chen, T.-L.: Integrating RFID with quality assurance system – Framework and applications. Expert Systems with Applications 36, 10877–10882 (2009) 13. Goodrum, P.M., McLaren, M.A., Durfee, A.: The application of active radio frequency identification technology for tool tracking on construction job sites. Automation in Construction 15, 292–302 (2006) 14. Feng, W., Jian, Z.: Consumers’ perception toward quality and safety of fishery products, Beijing, China. Food Control 20, 918–922 (2009) 15. Zhang, X.: Consumption trends and habits for fishery products in China. In: ASEM Aqua Challenge Workshop (2002)
Function Design of Township Enterprise Online Approval System Peng Lu1, Gang Lu2, and Chao Ding2 2
1 Tourism Department, Hebei Normal University, Shijiazhuang, Hebei, China School of Management and Engineering, Shijiazhuang University of Economics, Shijiazhuang, Hebei, China
[email protected],
[email protected],
[email protected] Abstract. Township enterprise is a kind of new economic organization that appeared under the special historical background in rural areas of China. Since 30 years of reform and opening-up, township enterprise has made a great contribution to the economic development of China with its unique development style and great vitality. Taking the place of the traditional manual way, township enterprise online approval system applies computer and network technology to realize the normalization and standardization of approval and administration of township enterprise. This paper first gives an introduction on the comprehensive function design of township enterprise online approval system, then makes an evaluation on the system, last points out the sphere of application of this system. Keywords: Township Enterprise; Online Approval System; Evaluation; Sphere of Application.
Since 1990s, human being entered into the Information-Dominated society, which is also called information society. According to the requirements of avoiding risk, moderate advancement and market operation, the management department of township enterprises should provide general business information services and agricultural enterprise information application services for characteristic industry, such as shoemaking, textile and garment, stone carving and stone processing, petrochemical industry, pottery and porcelain, hardware and electromechanical, tea, orange and so on. The agricultural enterprise application service system can provide township enterprises with integrated services in improving production efficiency, reducing cost and timely information gathering. At the same time, we should encourage the powerful and reputable information product enterprises to provide the peasants with both information services and business information in supplying and selling agricultural products. We should also strengthen supervision and crack down on price deception and the behavior of selling shoddy terminals so as to protect the legitimate interests of peasants. Therefore, we should build a government-dominating informationalized township enterprise constructing and organizing management institution and make an overall and long-term general plan for information development and a series of township enterprise e-business normative system to ensure the verity and health of the township enterprise information and the smooth channels of information collecting and D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 11–17, 2011. © IFIP International Federation for Information Processing 2011
12
P. Lu, G. Lu, and C. Ding
communicating and to strengthen the macro- guidance to the development of township enterprise e-business.
1 The Introduction of Township Enterprise Online Approval System The township enterprise business online approval system can achieve the standardization of township enterprise approval by using the computer and network technology instead of the traditional manual way to handle the application and approval of township enterprise. The administrative department of township enterprise acts as trade and economy commission in the villages and towns, as township enterprise administrative bureau or industrial promotion bureau in the county and also as a small-and-middle-sized enterprise bureau affiliated to industrial information office in the province. The responsibilities of the superior management departments are mainly
Fig. 1. The Overall Function Design of Township Enterprise Online Approval System
Function Design of Township Enterprise Online Approval System
13
for the interpretation and implementation of policies (referring to various kinds of national preferential policies and encouragement policies), calculating on numbers and examining and approving license for waste metal and coal marketing. The approval of township enterprises is handled by industrial and commercial bureau. If the enterprise is relatively large or with high technology content, the examination and approval shall be handled by development and reform bureau in the province, city and county first, and then get business license from industrial and commercial bureau. 1.1 Requirements of Township Enterprise Online Approval Function 1.1.1 Online Approval Module The online approval business includes the record keeping of the founding of township enterprise and its business change. The online approval functional module consists of the declarer’s bidding and inquiring function, assisting with input of villages and towns trade and economy commission and county township enterprises administrative bureau or industrial promotion bureau, industrial and commercial bureau’s three-level approval function and industrial and commercial bureau’s discipline inspection and supervision function. The online approval module, which is responsible for all the township enterprise approval business, is the core of township enterprise approval system. The necessary materials for township enterprises’ founding include enterprise name, domicile, business place, legal representative or legal person, business registration number, economic nature, organization type, scope of business, mode of operation, category of business, registered capital, number of employees and duration of operation and so on. The required certifying documents shall be uploaded in the forms of scanning or be filled in Word document format. It needs industrial and commercial bureau’s three-level approval in the online approval module. The industrial and commercial bureau’s three-level approval consists of preliminary reviewer’s examination and verification, competent business director’s examination and verification and competent business head’s examination and approval. If any level of the three-level examination and approval system doesn’t make any examination opinion, the examination and approval process will not continue. That is to say, the higher authority can only skim over the township enterprise’ application information and can't make any specific examination and approval opinion if the lower authority doesn’t express any opinion in accordance with the examination and approval process. The higher authority can make the decision of approval or not approval only when the low authority make the examination and approval opinion. This kind of process design can standardize the examination and approval process so as to avoid the skip-level examination and approval, which plays a significant role in the higher authority’s supervision to the lower authority. 1.1.2 Township Enterprise Information Management Module The township enterprise management module is responsible for collecting all the township enterprise information and managing the existing township enterprises. In this function module we can conduct such processes as certificate printing, information inquiry, township enterprise’s cancellation of registration, generating and exporting of information, the management of township enterprise business registration number and so on.
14
P. Lu, G. Lu, and C. Ding
1.2 Function Output of Township Enterprise Business Online Approval System Township Enterprise Business Online Approval System is composed of two sets of modules. One is Township Enterprise Approval System, including township enterprise set up approval system and township enterprise business change system; the other is Township Enterprise Database Management System, including inquiry system, report management system and certificate print system. Township Enterprise Online Approval System establishes relationships with terminal customers by relying on Internet. To ensure the safety of Township Enterprise Online Approval System and avoid attack and destruction of internet hackers, the system is divided into two web interfaces, one being logged in by natural persons and legal representatives who need transact township enterprise business through user name and password, another being used for approval of industrial and commercial bureau and assisting administrative department of township and county with inputting information through “dongle” password and user name and password. In addition, the system opens a discipline inspection and supervision window to make it easier to inspect and supervise so as to make the approval more transparent. 1.2.1 Township Enterprise Approval System The applicant logs in the application web interface of the Township Enterprise Online Approval System and input user name, password and random verification code successively, according to the indication of the computer, to enter the online approval application program of township enterprise. Then the above mentioned user name and password belong to the applicant who shall remember them carefully for future login and inquiry of relevant information. According to the indication of the computer, the applicant reads the provisions regarding to the establishment application of township enterprise in the Regulations on Township Enterprise and Implementation Rules. After deeply understanding the required documents, conditions and document format for application, the applicant shall click “Next” to fill in the township enterprise establishment form item by item carefully. During the process of filling in the above form, if the applicant fills in wrong contents or the contents filled in need to be modified, then he can click command of modification to fill in the form again. After that, he scans and uploads supporting documents according to the required format. Finally he clicks “Save” and submit the form. The filling contents will be transmitted to Information Center server of Administration for Industry and Commerce through network. The applicant, after receiving the documents returned from any approval level by industrial and commercial bureau, only needs to selectively modify the items not conforming to the provisions according to the suggestions. Then he can submit the documents and enter the approval link again. The purpose of adding the above function is to reduce the repeated input workload of the applicant and save application time. 1.2.2 Township Enterprise Business Change System The business change of township enterprise refers to the change of investor, legal representative, business operation site as well as change of its name. If any above items change, the township enterprise shall go to the original approval authority in time to go through the formalities for the change of the relevant business. In order to facilitate the operators, the township enterprise business change function module is designed in the Township Enterprise Online Approval System.
Function Design of Township Enterprise Online Approval System
15
Changing applicant uses user name and password to log in the application web interface of the Township Enterprise Online Approval System and enter the Township Enterprise Online Business Change System. Then he shall, in accordance with the indication of the computer, fill in the business change application form of township enterprise after reading the application documents needed to be submitted for the change. During the process of filling in application form, if the alteration applicant fills in wrong contents or the contents filled in need to be modified, then he can modify the contents or fill in the application form again. After that, he scans and uploads supporting documents according to the required format. After the saving and submission, the filling contents will be transmitted to Information Center server of industrial and commercial bureau through network. The applicant of township enterprise shall remember the user name for login and inquiry of relevant information in the future. The applicant for business change, after receiving the documents returned from any approval link by industrial and commercial bureau, only needs to selectively fill in the application form or modify the items not conforming to the provisions according to the suggestions. Then he can submit the documents and enter the business change link again.
2 The Evaluation of Township Enterprise Online Approval System The application of Township Enterprise Online Approval System is a useful exploration and an attempt at using modern network technology to approve township enterprise. It is an important measure for standardizing township enterprise approval management, and it indicates a higher stage of approval management of township enterprise. Meanwhile it is also a useful attempt at promoting Township Enterprise Online Approval System all over the country. 2.1 Convenient for Application The applicant can go through the application formalities of township enterprise at home through preparing the complete application documents in accordance with the requirements of Township Enterprise Registration and Record Stipulation. 2.2 Convenient for Supervision and Making Approval Procedure Opener and More Transparent Every step and link of online checking and approval is recorded in files, thus it is convenient for inquiry and for the leaders to supervise and examine; discipline inspection and supervision window is specially set in the online approval system, it is convenient for discipline inspection and supervision and by doing this, it strengthens the openness of approval. 2.3 More Standardized Approval Formality In the Township Enterprise Online Approval System, the traditional approval procedures and steps are edited into programs recognizable by the computer, the applicant
16
P. Lu, G. Lu, and C. Ding
only needs to submit step by step the necessary documents according to the conditions stipulated in the Township Enterprise Registration and Record Stipulation and the indication of the computer, then the industrial and commercial bureau can check and approve them step by step according to the programs of the computer, by doing this, it greatly reduces the human factors and non-standard operation in the traditional approval procedure. 2.4 High Safety and Reliability of the Whole Approval System The applicant, using his user name and password, logs in the application web page to submit application documents, while the administration authorities at different levels log in the checking and approval web page with their keys and user names and passwords to check and approve the documents. The method of using different web pages for application and approval respectively effectively avoids the attack and destruction by network hackers and ensures the safety and reliability of the whole approval system. 2.5 Greatly Improve Administrative Approval Efficiency The whole approval procedure, from the applicant’s logging in the network to the checking and approval of administrative authority of township enterprise, is very simple and swift and it lasts no more than half a day. The whole process of application, checking and approval is open and transparent, favorable for building a clean and honest government.
3 The Sphere of Application of Township Enterprise Online Approval System It is difficult for medium-and-small township enterprises to build a famous e-business platform since it needs supporting of technology and a large quantity of money. Township enterprise administrative authorities should build township enterprise private network and information sharing platform; construct a characteristic township enterprise information network system with sound system, fully functioning and highly practical and secure, which can make a link between the province, city, county, villages and towns and enterprises. And the network can also become a promoting platform to domestic and international market and in this way promote township enterprises e-business actively to intensification, economizing and efficiency.
4 Conclusion The township enterprise online approval system belongs to the areas of e-government, and the main objective is to provide a convenient and effective way of doing business for the applicants. How to save the application time and cost of user and put forward a realistic model of the online approval system would be the basis for the township enterprises to transfer to the network. Using decomposition of structured analysis method step by step, the paper proposes a functional model of township enterprises
Function Design of Township Enterprise Online Approval System
17
online approval system. According to the characteristics of township enterprise business and combining with the traditional method of township enterprises business, the paper puts forward a kind of functional model of different functional modules and roles, which relies on the contacts between the internet network and end users. The functional model allows the user to shorten the application time to the minimum and satisfies the needs of users. It achieves standardization of township enterprise approval and management, thus reduces processing costs. It will also accept social supervision, so that the efficiency can be improved, and it serves the community better. In order to make the model functions well, it still needs to combine with the actual function of township enterprises approval, so that the function of the system can be further close to the actual environment.
References 1. Aharon, K.: Phases in the Rise of the Information Society. Info (2000) 2. General Office of the CPC Central Committee, General Office of the State Council: National Informationization Development Strategy 2006—2020. ZBF (2006), http://www.cnii.com.cn/20050801/ca350966.htm 3. Zhang, J.: Practical Exploration of Online Approval System. Journal of Ningbo Radio & TV University (2007) 4. Zhang, J.: Design and Realization of Online Approval System. Journal of Lujiang University (2005)
Application of GPS on Power System Operation Chunmei Pei1, Huiling Guo2, Xiuqing Yang1, Bin He1, Wei Liu1, and Xuemei Li1 1
Beijing Vocational College of Electronic Science and Technology, Beijing, China 2 China University of Mining Technology (Beijing), Beijing, China
[email protected] ,
Abstract. Applying GPS positioning and navigation technology to power systems will realize precise navigation of power equipments, improve power system automation of routine works, and enhance working efficiency. In case of emergency, rapid fixing arrangements can be implemented through monitoring and command platform. With advance communication technology, the realtime video can be transferred to the experts all over the world for remote joint consultation. Keywords: Communication Terminal Equipment, Position and Navigation.
,GPS Technology, Power Transmission
1 Introduction After years of development, GPS system has been extensively spread out from mainly military usage to civilian ones. Now more and more GPS terminal, such as PND (Portable Navigation Device), CND (Car Navigation Device), GPS Cell Phone, became popular in people’s daily lives. GPS-related applications of various industries have also gradually growing and many industries have their own GPS applications. At present, most technologies of modern Power Enterprises have met international standards, but the methods of inspection and location are at a lower level, still relying on manpower. The equipments of Power Enterprise are diversiform and widely distributed, so it is a fact that many inspection and repairing staff are not familiar with the equipments’ accurate positions. The traditional method is that senior staff lead the way for young people, which is a waste of manpower. Particularly in the accident emergency, the traditional method cannot ensure technical personnel to arrive at the scene in time. GPS positioning and navigation system will replace the traditional method with high-tech satellite navigation, which is an ultimate solution to the current problems.
,
2 The Advantage of GPS Positioning and Navigation System in Electrical Power System 2.1 Improvement in Efficiency of Daily Operation and Maintenance China’s territory is vast. The number of electrical power equipments is big, and they’re widely distributed. Lots of devices are deployed in remote areas, which brings D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 18–22, 2011. © IFIP International Federation for Information Processing 2011
Application of GPS on Power System Operation
19
much inconvenience for inspection and checking. It’s difficult to find the equipments. The traditional method is to train the new generation of staff by senior ones. This approach is neither scientific nor efficient, especially for large-scale activities (such as the Olympic Games), which need regional cooperation. The GPS positions of all equipments are entered into the map data, so everyone can easily find the destination with a GPS device. Labor cost is saved and the efficiency is significantly improved. 2.2 Improvement in the Ability of Handling Accidents Accidents can not be completely avoided today. In addition to active prevention, a quick solution is particularly important after it happens. Quick and accurate arrival at the scene with a precise navigation device could avoid great loss of the country. Meanwhile, with advanced wireless communication systems, real-time videos of the accident are transmitted to the experts all round the country for remote diagnosis, which can greatly enhance the ability to handle accidents.
3 The Benefits of GPS Positioning and Navigation System in Electrical Power System 3.1 Considerable Economic Benefit Power failure could cause national economic disaster. A malfunction may cause economic losses from a few hundred million to more than a billion dollars. (Northeast blackout in United States, 2003, according to the USA and Canada Joint Investigation Team’s published report in Dec 5,2003, was the most serious in U.S. history to make a total blackout affecting about 50 million citizens. During the two day power outage, plants shut down and companies stopped businesses caused 4 to 10 billion U.S. dollars losses.) When an emergency occurs, every second can be extremely valuable. GPS positioning and navigation system can realize fast positioning, rapid troubleshooting, and hence avoiding economic losses as much as possible. Meanwhile, the reservation of the guides can be avoided in daily operation and maintenance, even in large cross-regional operations. The labor cost is greatly saved. 3.2 Large Social Benefits Electrical power is closely related to everyone’ life in civil society. In any city, power failure is an inconceivable disaster. Several large-scale power outage caused by the accident in history gave local people painful memories. Thus, with GPS positioning and navigation system, the ability of troubleshooting is enhanced for Power Enterprises, which has a very significant impact on people’s livelihood and social stability.
4 The Main Functions of GPS Positioning and Navigation in Electrical Power System 4.1 Positioning of Transmission Towers, Substations and Offices The locations of transmission towers, substations, offices and so on, are preset in GPS devices, as in Fig. 1. With an electronic map in GPS devices, after positioning by
20
C. Pei et al.
GPS satellites, the route can be automatically calculated by directly clicking on the destination. The route from the start point to the destination can be reasonably planned and tracked, as in Fig. 2. 4.2 Dynamically Addition or Subtraction of Location Information According to Requirement The system employs open data structure, which makes it convenient to add or remove location information. When new equipments are added, or some equipment is out of
Fig. 1. To preset location information
Fig. 2. To start navigation
Application of GPS on Power System Operation
21
date, the user can easily make the change by him/herself. The system also realizes data sharing of address book and navigation path with different devices. Crossregional cooperation can be achieved by simply integrating location information of different areas. New devices are no longer required. 4.3 Real-Time Monitoring, Improving Management Efficiency The GPS navigation device receives GPS satellite signals, automatically positions, and sends the location information in the forms of SMS or data (GPRS / 3G, etc.), to the master control center, via built-in wireless modules, as shown in Fig. 3. The master control center receives the information, extracts the location information, and dynamically displays the longitude, latitude, speed, status, etc., of the vehicles on the electronic map, as shown in Fig. 4. By integrating the data collected, the corporation can find the most appropriate operating fashion, avoid waste, and save cost. By digging deeper into the data, the analyzer can provide the most authentic and reliable reference to the management team, to make more opportunities.
Fig. 3. GPS Monitoring System
4.4 Combined with Advanced Network, to Enhance Emergency Response Capabilities When dealing with urgent accidents, the monitoring platform can accurately obtain the distribution of vehicles and personnel, and carry out overall arrangements. At the same time, GPS positioning and navigation system not only can guide staff to the scene quickly, but can communicate with supervisors via wireless communication capabilities in time, to obtain the correct commands. When facing with complex problems, live scene video can be sent to the master-monitoring center via advanced wireless network (GPRS/3G, etc.). Experts from different regions can participate in the multi-party consultation to diagnose and resolve the problem in time, saving the loss.
22
C. Pei et al.
Car No.1 80km 52.213918,7.12532
Fig. 4. GPS system monitoring interface sketch
5 Conclusions Electricity is closely related to people's lives. Power Enterprises are taking on more social responsibilities than other general ones. It is Power Enterprises’ duty to keep the power grid working stably and regularly. GPS positioning and navigation system, which relies on advanced technologies and is integrated with modern Internet applications, will greatly reform the way of inspection and routine work, thus improving working efficiency. It can also play a valuable role in emergency. As GPS positioning and navigation system is widely used in Power Enterprises, it is no doubt that the whole power grid’s modernization and technology level will be improved. It will make a positive contribution to the society.
Reference 1. Zhu, K.: GPS Application in Power System. J. Computer & Digital Engineering, Beijing (2007)
Greenhouse Temperature Monitoring System Based on Labview 2
Zhihong Zheng1, Kai Zhang1, and Chengliang Liu 1
School of information and control, Nanjing university of information science & technology, Nanjing Jiangsu, China, 210044 2 School of mechanical and power engineering, Shanghai Jiao Tong University, Minhang Shanghai, China, 200240
[email protected],
[email protected],
[email protected] Abstract. The environmental temperature plays an important role in the growing crops. How to make use of computer technology to realize automatic control ambient temperature of greenhouse is a hot issue in the intelligent agriculture. The control of environmental temperature in modern greenhouses is collected and analyzed by monitoring system of greenhouse environment based on Labview software and wireless communications technology. NRF24L01 wireless receive-send model makes temperature collection come true. Moreover, managing computer makes data wireless communications become possible. Over- temperature alarm information is transmissioned to user timely. Finally system interface is designed on the Labview software platform. Keywords: temperature, intelligent agriculture, labview, temperature sensor.
1 Introduction Facility gardening is a kind of production with the bad-effect condition in which the crops (flowers, fruit trees and vegetables) don't tend to grow normally in the cold or hot season .Thus people must utilize heat preservation, cold-proof, temperature reduction, defense and equipments in a man-made way to create an environment that favors the crop's growth without the effect of climate change[1]. It is a comprehensive greenhouse technology, which meets the demand of ecological condition in plant including light, temperature, water, gas, soil and nutrition. Planting in the different seasons is able to yield high vegetables and fruits production in good quality. Comparing with foreign countries, overall technology ability of national facility gardening is poor due to the later start and shorter development time, and therefore environmental regulation should be enhanced. Diverse facility structure, automatic production management, mechanized operation, production intensification are the typical features in Holland, America, Japan, France and Israel. In those countries, modern industry, high technology and advanced management equip agriculture. We utilize computer and information technology to realize intelligent management of the temperature. The system adopts Labview software, wireless communications technology and GSM technology to realize temperature auto- monitoring and alarm. It is a good way to facility gardening automatic management. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 23–29, 2011. © IFIP International Federation for Information Processing 2011
24
Z. Zheng, K. Zhang, and C. Liu
2 Wireless Communications on NRF24L01 It is good for monitoring environmental from several spots due to large gardening scale, especially in a sunny exposure and at vents. Outside environment largely influence the temperature. So we should set more spots to collect and pay special attention to the important spots. It is the best to adopt wireless communications and use lower powered chip to timely send collected temperature to managing computer due to difficult allotting the positions. In this thesis, NRF24L01 wireless receive- send pattern as wireless communication equipment and set 6 sensors (all are 18b20) to attache to it. The place is that respectively two positions at vents, middle and side in the greenhouse. The collected environmental temperature is sent to store in the computer by NRF24L01 pattern in time. 2.1 NRF24L01 Chip Introduction NRF24L01 chip adopts 2.4 Ghz global ISM band comply with 126 multi- spots and skip band communications requirement. Build- in CRC hardware error detecting and one to many spots communications requirement. Module is able to set up address. If only receiving this computer address, the data can be sent out(offer interrupt directions). Directly connected with single chip to use, programming is very convenient. Receive-send module is categorized to Enhanced ShockBurstTM, ShockBurstTM and Direct ways, which is made up to device configuration[5]. Four parts as bellows: Data width: declaring data occupied decimals in the radio frequency database. It renders NRF24L01 to tell data of receive-send database from CRC code; Address width: declaring address occupied decimals in the radio frequency database. It renders NRF24L01 to tell address from data; Address: receive data address, address of Passage 0 to Passage 5; CRC: yield CRC check code and decoding If use CRC technology within NRF24L01, CRC checking code should be used in the configuration( CONFIG’s EN_CRC). Send and receive the same protocol. Under the Enhanced ShockBurst TM pattern, use first-in and first-out stack area, data is sent to micro controller with low speed. But it can save more energy under high speed. Therefore, using low speed micro controller get high sending ratio as well. All the high speed signals processing in the chip under ShockBurstTM receive send pattern, NRF24L01 automatically process character and CRC checking code. When sending the data, character and CRC checking code is added. EC is high under sending pattern. It will take 10us to send over. 2.2 Data Receive- Send Process of NRF24L01 NRF24L01 adopts Enhanced ShockBurstTM pattern to send,it is shown as bellows: send the address and data to NRF24L01 as time sequences; configure CONFIG register and make it access to sending pattern; micro controller put CE higher(at least 10us), activate NRF24L01 and send Enhanced ShockBurstTM
Greenhouse Temperature Monitoring System Based on Labview
25
EnhancedShockBustTM sending has four steps: (1) supply electricity to ratio front; (2) Data packing ( adding character and CRC checking code ); (3) sending data package with high speed ; (4) send it over, NRF24L01 enters idle state. Sending BYTES data procedure via NRF24L01 pattern are as follows: uchar SPI_Write_Buf(BYTE reg, BYTE *pBuf, BYTE bytes) { uchar status,byte_ctr; CSN = 0; status = SPI_RW(reg); for(byte_ctr=0; byte_ctr //set receive-send way via SIM card AT + CMGF = 1 < CR >
//set Text pattern
AT + CMGS =″15951775730″< CR > // send one short message, 15951775730 is target address Warning!
// SMS content
AT + CMGR = 6 < CR > // read receiving SMS, suppose if new SMS’s postion in the SIM is 6 + 861595177573 is mobile service center number in Nanjing area. The side double quotation mark should be sent. < CR > is enter mark. Each AT instrucion is end up with enter mark; Warning! That’s the SMS content; is CTRL + Z,SMS should be ended up with < ^Z >. Last line is readable SMS instructions, in which 6 is SIM index number. Once that instrucion is sent, GSM pattern sends bellow informations back from UART interface: + CMGR :″REC UNREAD″,″+ 8615951775730″,″19/ 05/ 10 ,15 :54 :00 + 00″Warning! It is easy to read out sender’s mobilephone number,sending time and SMS conetent. The system writes AT intructions into virtual serial, which comes out by Labview software. There are 6 AT instructions to write down by sequence. So each time we respectively write down different instrutions with condition structure in the circular struction. Procedures are as follows:
Fig. 2. The program is to write AT instructions on Labview
4 Making Use of Labview Software Developing System Platform Labview is a kind of developing environment of graphic program language. It is widely accepted by the industry and education and lab, which is considered as a standardized data collecting and equipment control software. LabVIEW integrates the hardware in the GPIB, VXI, RS-232 and RS-485 agreement and the entire functions from data collecting card communications. It also contains standard library function that is easy to apply for TCP/IP, ActiveX, etc. It makes user rapidly build self data
28
Z. Zheng, K. Zhang, and C. Liu
Fig. 3. The above waveform shows temperature change curve within one day. There are 48 points in the X, one temperature is collected per half an hour. It directly shows temperature change conditions in one day. But six numbers below waveform shows current temperature by 6 sensors. Obviously, the value of Sensor 1 and Sensor 3 at the center position is lower. But the value of Sensor 5 and Sensor 6 at the corner is a little bit higher.
and analyzed system, and then system over- temperature alarm information is sent to user via serial writing functions. TDMS data storage formats is mathematics model specially designed for data storage. It’s characterized as steady read data API and self- configured data management tools that are used to manage data. The collected temperature each time is stored as binary in the TDM format folder. The value from the sensor can be find out by sequence. On the system interface, value can be traced through replaying the data and time input. Alarm light will turn to red to remind user that the current temperature has exceeded the defined figure. Thermometer shows the maximized temperature value. Below is monitoring interface.
5 Conclusion Through the analysis of the monitoring results, it turns out that,when the change of environmental temperature is large in one day, the hysteresis of temperature adjustment mainly manifested in space. In the middle position near the thermostat, sensor detects the environment temperature unchanged, equal to the set temperature. But in the corner of the greenhouse external environment, the detected temperature is affected by the temperature of external environment, there exist certain volatility. Therefore, it is necessary to install the adjusting devices at the position where the temperature is vulnerable effectted by the external environment temperature, it can reduce the influence of temperature change for crops. The control of environmental temperature in modern greenhouses is collected and analyzed by monitoring systems of facility gardening environment based on Labview software and wireless communications technology. NRF24L01 wireless receive-send model makes temperature collection come true. Moreover, managing computer makes data wireless communications become possible. Over- temperature alarm information is transmitted to user timely.
Acknowledgment This paper is sponsored by National High Technology Research and Development Program 863 (NO:2006AA-10A301).
Greenhouse Temperature Monitoring System Based on Labview
29
References 1. Yu, H., Zhang, Y., Sun, R.: LabVIEW-based research of remote monitoring system for greenhouse. Agricultural and Machinery Study, 75–77 (2004) 2. Qiu Shi Science and Technology: The Navigation of Typical Module Design by MCU, pp. 194–202. Posts & Telecom Press (2004) 3. Cheng, X., Zhang, Y.: LabVIEW 8.2 Programming from Entry to Maste, pp. 312–315. Tsinghua University Press (2007) 4. Qi, F.: Software Implementation of Intelligent Control System of Greenhouse environment based on Labview, pp. 1–2. Zhejiang University (2004) 5. Shenzhen Yun Jia Technology Co., Ltd.: NRF24L01 Manual, pp. 6–10 (2008) 6. Wu, B., Liu, X., Wu, M.: Reserch of GSM-based Universal Remote Alarm Controller. Computer Engineering and Application, 92–94 (2007)
Image-Driven Panel Design via Feature-Preserving Mesh Deformation Baojun Li1, Xiuping Liu2, Yanqi Liu2, Ping Hu1,*, Mingzeng Liu1, and Changsheng Wang1 1
School of Automotive Engineering, Faculty of Vehicle Engineering and Mechanics, State Key Laboratory of Structural Analysis for Industrial Equipment, Dalian University of Technology, 116024 Dalian, China 2 School of Mathematical Sciences, Dalian University of Technology, 116024 Dalian, China {bjli,xpliu,pinghu}@dlut.edu.cn
Abstract. In this paper, we propose an image-driven 3D modeling technique for rapid panel design. Our semi-automatic approach is based on template technique and mesh volume deformation controlled by a special cage. We designed our modeling system to be interactive in 2D, automating the process of shape generation while relying on the user to provide image samples. Once a parametric model template is given, using the contour extracted from images, the new control cage corresponding to mesh models generated. Then the geometry of new panel is automatically recovered from the deformable template model. Our system also allows the user to easily reconstruct other 3D objects in a similar manner, such as realistic-looking plant modeling from images. We show realistic reconstructions of a variety of panels, automobile shapes and demonstrate examples of plant editing. Keywords: Image-driven, Mesh deformation, Panel Design, Deformable template.
1 Introduction Nowadays polygon meshes are widely used in both geometric modeling and finite element analysis fields. Mesh deformation is useful for providing various shapes of meshes for CAE tools, especially in very early phases of conceptual automotive design. Most of 3D objects are largely dominated by a few typical features which include contours, even and engineered meanings. Thus, to create 3D models with new appearances by processing and reusing the existing models is becoming an extremely important way to ease the efficiency problem of geometric design in computer-aided design and computer graphics. Space deformations (volume deformation) are to this day the method of choice for shape deformation due to its independence of surface representation[1]. A space deformation is defined via a (usually simple) control cage or grid; user-defined deformation of this object is interpolated to the 3D space and evaluated at the input surface *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 30–40, 2011. © IFIP International Federation for Information Processing 2011
Image-Driven Panel Design via Feature-Preserving Mesh Deformation
31
points. Space deformations are popular since they can handle various object representations, including parametric surfaces, polygonal meshes with multiple connected components, etc. In addition, space deformations are simple to implement, and they are highly efficient and robust, because the cost of the deformation is mainly dependent on the complexity of the control object and not on the deformed shape. Early space deformations used lattices as control objects, and then had also been explored as FFD[2], EFFD[3], and DFFD[4] etc. Later work proposed the use of so-called cages as control objects for shape deformations. Typically, the cage is a very coarse and offsetted version of the input shape. Various coordinate functions have been designed to carry over the deformation of the cage to the entire space, such as mean-value coordinates[5], harmonic coordinates[6], Green coordinates[7]. However, cage-based deformation schemes in references [5][6][7] which preserve differential properties, so far, cannot support direct manipulations very well. Cages generation and its manipulations are more complicate. Thus, in this work, for more adaptive for different kind of FEM meshes and easy to edit in 1D, we adapt modified DFFD [4] method as our main deformation technique. The core techniques of the reuse of existing 3D mesh models using mesh deformation, the key property of mesh editing is interaction techniques and more input examples, such as images, 2D sketches. Masuda et al. proposed a combination method to manually specify varying surface stiffness for panel design[10], [11]. Gal et al. introduce a so called iWIRES, a novel approach based on the argument that man-made models can be distilled using a few special 1D wires and their mutual relations0. In this paper, a simple 1D editing method based on symmetric projection is developed, which deals with the control box of DFFD easily and can support the image examples very well. In this paper, we introduce an image-driven mesh morphing framework for rapid design of the automobile panels, and also generalize this method to other applications, such as realistic-looking planting modeling from images. Section 2 presents the whole pipeline of the method. Then the detailed algorithm of the mesh deformation and contour extraction from images are introduced in Section 3. Section 4 gives the numerical examples and discussion.
2 Overview of Image-Driven Framework For production of high quality panels in a short period of time, the more freely and rapid design technique is widely studied in the past few years. Inspired by IWires method [1] and image morphing[12] method, our main motivation is to generate new models rapidly from images; meanwhile, users can edit the shapes in 1D. In this section, we will introduce our framework of the image-driven panel design based on mesh deformation. Fig. 1 shows the whole process of our method proposed in this paper. There are three main parts of this framework, which include contour extraction from images, generation of initial control box CB0 and new control box CB1, deformation and related verification via CAE software and aesthetics respectively. The pipeline of the framework illustrated in Fig. 1 is given briefly as follows. Pre-processing (Control box generation) In the pre-processing phase, the deformable mesh template M should be chosen and constructed properly according to needs of the practical panels. The first step of the
32
B. Li et al.
CB0 generation algorithm is to compute the bounding box of the initial dense mesh model M by principal component analysis (PCA); the so constructed bounding box captures the major geometric shape of the mesh M. For the simplicity of editioninteraction, we consider the main contour of the model via projection mapping, see Fig. 2. Then the corresponding contour and the control box CB0 associated with DFFD method should be constructed automatically.
Fig. 1. Flowchart of our method
(a)
(b)
(c)
(d)
Fig. 2. (a-d) Control box generation
Image-Driven Panel Design via Feature-Preserving Mesh Deformation
33
Fig. 2(a-d) illustrates the construction process of the control box corresponding to the given model in Fig. 2 (a). More details are described in Section 3 and Section 4. New control box constructed from images For the purpose of the image-driven mesh deformation, the key point is to create new control box from input image examples. Moreover, there are two core steps should be done carefully, which are the contour extraction from images with high precision and subsequent reconstruction of the new control box respectively. This stage will be described particularly in one full section due to its complicity with many techniques, see Section 4. Panel design via mesh deformation With the deformable template and the control boxes CB0 and CB1 are given, the next task is to generate the new panel or model using mesh deformation. The final mesh models are generated to meet requirements of panel CAE analysis, such as crash, NVH, durability and formability. Thus, a general deformation method which is independent of surface representation will be developed. In this paper, FFD method is used. In the next section, the mesh deformation techniques are described in detail.
3 Design via Mesh Deformation The rapid design of panels or carbody shapes is constrained by the conflicting requirements of multiple objectives. For example, designers need a more conventional CAD tool to modify the original models to obtain the shape changes required and time constants necessary for these changes. However, the final design is determined by CAE engineers and designers to verify the artistic and reliability of the shapes. State-of-the-art commercial CAE software, such as LS-Dyna, Nastran, Abaqus, PamCrash, Fluent, etc. are all based on meshes. Thus, this paper focuses on the platform which enables the user to rapidly change an existing FE / CFD mesh into a new target shape without having to redraw it in the CAD system. In this section, the mesh deformation method used in this paper will be introduced in detail. 3.1 Pre-processing Firstly, in order to rapidly generate a panel model, it is important that the suitable deformable template should be chosen and constructed. Thus, a series of panels or carbody database is constructed to meet the requirement of the different products. Once the template model was constructed, the following step is to generate a corresponding control box, which can design new shapes via space deformation method. In this paper, for the sake of edition simplicity, the 2D edition approach will be adopted. Now we introduce our main methods used in this work in brief. The automatic CB generation steps are described as following: 1) Compute the principle axis of the template model by Principal Component Analysis (PCA)[17]; 2) Rotate and project the model into the principle plane, then obtain the contour polygon, see Fig. 2;
34
B. Li et al.
3) Automatic generation of the control polygon feature points from the contour; see Fig. 2b; 4) Semi-automatic generation of control box CB0 by tensor production to the principle axis, see Fig. 2 c-d. It is noted that the suitable CB0 should be construct carefully, which affects the deformable template and the final design. Meanwhile, control boxes CB0 and the corresponding CB1 also provide the feature-preserving of the template through space deformation method. In this work, we use the projection method in order to simplify the user-interaction process, that is, designers can modify the shapes in 2D principle plane but in the complicated 3D model directly. Taking the advantage of 2D editing, our method can generate a new model just from an input image or image examples, which is so called image-driven panel design method. Fig. 3 illustrates a new control box generated by our method.
Fig. 3. New control box generation by projection
3.2 New Control Box Generated from Images Due to editing the control box CB0 in 2D while in 3D directly, our method can generate a new control box from a new contour with some constraints. So, an image-driven modeling method is proposed based on the mesh space deformation method. This part is of the most importance in this framework, and will be given in Section 4. 3.3 Mesh Deformation In order to make use of surface-based techniques for deforming automobile sheetmetal panels, Masuda et al.[10][11] develop the soft and hard constraints on the mesh and propose a framework which can preserve the form features of the sheet-metal panel while deforming the model. Huang et al. [13] proposed a morphing method with feature-preserved for panel design, which use the DFFD method on NURBS surface directly. However, as described in Section 3, the rapid panel design method should provide different meshes to satisfy the need of subsequent CAE analysis. Thus, in this work, the mesh space deformation theory is used rather than the surface-based method.
4 New CB1 Constructed from Images In this section, we introduce the detailed techniques to reconstruct new control box CB1 which shares the same topological structure as the original one CB0, see Fig. 4.
Image-Driven Panel Design via Feature-Preserving Mesh Deformation
35
Fig. 4. Pipeline of CB1 reconstruction from an image
4.1 Contour Extraction from Images First of all we introduce how to extract the exact contour information from an ideal panel image which is the key point to generate new control box. As shown in Fig. 4, the input images should be pre-processed to obtain a contour with high precision firstly. There are two necessary operations to improve the extracted result in the pre-processing phase, which are image resizing and image smoothing respectively. In this work, for higher effencicy, the bi-linear interpolation method is used to resize an image for ideal; In order to obtain an initial contour from the input image, Guassian smoothing filter is used. Fig. 5 or an imput image example, and the Fig. 5 obtained after the pre-processing stage. As the pipeline shown in Fig. 4, the following step is to extract the contour from the image, as shown in Fig. 5b-c. The improved gradient vector flow snake (GVF) [14] method is used in this paper. Now we introduce this method in brief. A GVF model is the vector field
v ( x, y ) = ⎣⎡u ( x, y ) , v ( x, y ) ⎦⎤
,
which minimizes the energy functional
ε = ∫∫ μ ( u x2 + u y2 + vx2 + vy2 ) + ∇f
2
2
v − ∇f dxdy .
This variational formulation follows a standard principle, that of making the result smooth when there is no noisy data. In particular, when | ∇f | is small, the energy is dominated by sum of the squares of the partial derivatives of the vector field, yielding a slowly varying field. On the other hand, when | ∇f | is large, the second term dominates the integrand, and is minimized by setting v =| ∇f | . This produces the desired effect of keeping v nearly equal to the gradient of the edge map when it is large, but forcing the field to be slowly-varying in homogeneous regions. The
36
B. Li et al.
(a)
(b)
(c)
Fig. 5. Contour extraction from an image
parameter μ is a regularization parameter governing the tradeoff between the first term and the second term in the integrand. This parameter should be set according to the amount of noise present in the image (more noise, increase μ ). 4.2 Post-Processing of Contour In the above subsection, the initial contour is obtained from an image, as illustrated in Fig. 4a-c. However, the initial contour cannot match with the contour of original control box CB0 very well. There are two necessary operations to solve the problem, which are contour assessment and feature point assignment respectively. Contour assessment Once the template model is given, through the pre-processing operation in Section 3, a constant original contour is obtained. In order to generate a final control box CB1 with the same topology and features as CB0, the original contour and contours extracted from images should be assessed exactly. Fig. 6 shows the geometric assessment result for a given carbody model, and Fig. 6b illustrates assessment result of the contour shown in Fig. 4c. It is should be noted that this step can be an automatic process for a given specific panel case.
(a)
(b) Fig. 6. Geometric assessment of contours
Feature point assignment There is another important point of contour post-processing, i.e., feature point assignment. It is obvious that there are no any apparent relationships between the original contours and contours extracted from images. Thus, to bridge this gap, we should assign some specific feature points on the contours which determine the panel shapes and main corresponding engineering meanings. In this paper, we assignment the
Image-Driven Panel Design via Feature-Preserving Mesh Deformation
37
Fig. 7. Coding the feature points
feature points semi-manually. As shown in Fig. 7. Coding the feature points, the coding of the feature points for contour in Fig. 4c is given. 4.3 Reconstruction of New Control Box CB1 By extraction from images and suitable post-processing, a coded contour with geometric assessment is obtained. In this step, a new control box CB1 with the same topology with CB0 will be constructed using the above contour information. The geometric transformation between the contour from image and the contour associated with CB0 is used. Due to the difference of geometric information between two contours, we adjust the control polygon of the contour extracted from image, according to the feature calibration and geometric information of the original contour. Fig. 8 illustrates an adjustment for the contour shown in Fig. 6 (in blue), and the red one denotes contour associated with the template. After this operation, the control polygon and contour of CB1 are obtained.
Fig. 8. Adjustment of the control polygon
Fig. 9. The final contour of the CB1
In order to construct a new control box CB1 with high precision and the same topology with CB0, a resampling operation with high precision is necessary. However, the contour extracted from an image directly usually has a low-resolution. In this paper, cubic spline interpolation method[16] is used for the initial extracted contour.
38
B. Li et al.
Then the final contour of the CB1 is obtained by resampling from the interpolated spline curve with the same number of points with CB0. Fig. 9 shows the final contour of the CB1 corresponding to Fig.5a-c.
5 Implementation and Numerical Examples We have applied our method to a large variety of panel and carbody models, and obtained very ideal results. All models are computed on a double 2.80-GHz PentiumR (2G RAM) machine, using VC++ 6.0 and KMAS/COMX development platform. In this section, three images of cars from Internet are given, and the associated carbodies are computed rapidly by our framework. Fig. 10 is the template model used in this paper, and the final corresponding designs are given as following, such as Fig. 11-13.
Fig. 10. The template carbody model in this paper
Based on the template model shown in Fig. 10, a new carbody design from image in Fig. 5 is obtained as following figures using the new control box illustrated in Fig. 3 and Fig. 9. Fig. 11 bottom shows the final FEM mesh of the new design, and the others shows the shading one from different views.
Fig. 11. The final design corresponding to Fig.5-9
We also provide two carbody designs from images directly using our method, see Fig. 12 and Fig. 13.
Image-Driven Panel Design via Feature-Preserving Mesh Deformation
39
Fig. 12. New design generated using our method
Fig. 13. New SUV design generated using our method
6 Conclusions In this paper, we proposed a framework for rapid panel design based on mesh space deformation. Furthermore, we also explored the approach to create any other kinds of man-made, engineered objects and realistic shapes. This approach generates so many different kinds of shapes due to image example input which is abundant from internet. Numerical examples show that our framework is effective and able to reconstruct visually pleasing objects. The final output finite element meshes can be utilized by CAE software. It should be noted that this study has examined only for auto-body design in the conceptual design stage. For the future work, we will improve current contour-extraction method from images, in order to obtain the feature information with higher precision. A more automatic registration algorithm to reconstruct new control boxes CB1 from images will be also considered. Furthermore, a more parametric framework will be constructed to generate panels or carbody self-adaptively to meet designer’s needs. More applications under this framework will be also considered.
Acknowledgement This work was funded by the Key Project of the NSFC (No. 10932003, u0935004), NSFC (No. 60873181), “863” Project of China (No. 2009AA04Z101), “973” National Basic Research Project of China (No. 2010CB832700) and the Fundamental Research Funds for the Central Universities. The model in Fig.10 is provided by DEP.
References [1] Gal, R., Sorkine, O., Mitra, N.J., Cohen-Or, D.: iWIRES: An Analyze-and-Edit Approach to Shape Manipulation. ACM Trans. Graph. 28(3), 1–10 (2009) [2] Sederberg, T.W., Parry, S.R.: Free-form Deformation of Solid Geometric Models. In: Proc. of ACM SIGGRAPH 1986, pp. 151–160. ACM, New York (1986)
40
B. Li et al.
[3] Coquillart, S.: Extended Free-form Deformation: A Sculpturing Tool for 3D Geometric Modeling. In: Proc. of ACM SIGGRAPH 1990, pp. 187–196. ACM, New York (1990) [4] Hu, S., Zhang, H., Tai, C.: Direct manipulation of FFD: Efficient explicit solutions and decomposable multiple point constraints. Visual Computer 17(6), 370–379 (2001) [5] Ju, T., Schaefer, S., Warren, J.: Mean Value Coordinates for Closed Triangular Meshes. ACM Trans. Graph. 24(3), 561–566 (2005) [6] Joshi, P., Meyer, M., DeRose, T., Green, B., Sanocki, T.: Harmonic Coordinates for Character Articulation. ACM Trans. Graph. 26(3), #71 (2007) [7] Jiang, N., Tan, P., Cheong, L.F.: Symmetric Architecture Modeling with a Single Image. ACM Trans. Graph. 28(5), 1–8 (2009) [8] Tan, P., Zeng, G., Wang, J., Kang, S.B., Quan, L.: Image-based Tree Modleing. ACM Trans. Graph. 26(3), 87–93 (2007) [9] Lipman, Y., Levin, D., Cohen-Or, D.: Green Coordinates. ACM Trans. Graph. 27(3), 1–10 (2008) [10] Masuda, H., Ogawa, K.: Application of Interactive Deformation to Assembled Mesh Models for CAE Analysis. In: ASME Int. Design Engineering Technical Conferences (2007) [11] Masuda, H., Yoshioka, Y., Furukawa, Y.: Preserving Form Features in Interactive Mesh Deformation. Computer Aided Design 39(5), 361–368 (2007) [12] Chen, L.L., Wang, G.F., Hsiao, K.A.: Affective Product Shapes through Image Morphing. In: Proceedings of the International Conference on Designing Pleasurable Products and Interfaces, pp. 11–16. ACM, New York (2003) [13] Huang, Q., Li, B.J., Liu, M.Z., Bao, J.R.: Feature-preserved Morphing Method for Panel Design. Mathematical and Computer Modelling 51, 1417–1420 (2010) [14] Xu, C.Y., Jerry, L.P.: Snake, Shapes and Gradient Vector Flow. IEEE Trans. on Image Processing 7, 359–369 (1998) [15] Xu C.Y., Jerry L.P.: Gradient Vector Flow Deformable Models. In: Handbook of Medical Imaging, pp. 159–169 (2000) [16] Wang, R.H.: Numerical Approximation. Higher Education Press, Beijing (1999) [17] David, L.: Linear Algebra and Its Applications. Addison-Wesley, New York (2000)
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber on the Cooling Rate during Vacuum Cooling Tingxiang Jin*, Gailian Li, and Chunxia Hu School of Mechanical & Electricity engineering, Zhengzhou University of Light Industry, 5 Dong Feng Road, Zhengzhou 450002, Henan Province, P. R. China Tel.: +86-371-63556785
[email protected] ℃
Abstract. The temperature of vapour-condenser below 0 and the final pressure in the vacuum chamber below 0.61kPa during vacuum cooling were experimentally analysed in this paper. The temperature of vapour-condenser, -2 , -35 , -39 and -71 , and the final pressure in the vacuum chamber, 0.3kPa, 0.4kPa, 0.5kPa and 0.61kPa, were chosen. The experimental results showed that the cooling rate varies with the temperature of vapour-condenser and the final pressure in the vacuum chamber. Water vapour becomes the frost on the surface of vapour-condenser when the initial temperature of vapour-condenser is below 0 , which is helpful to trap water vapour for vapour-condenser. In addition, the formation mechanism of frost at the surface of vapour-condenser was analysed in this paper. The cooling time for vacuum cooling can be reduced when the final pressure in the vacuum chamber varied from 0.4kPa to 0.61kPa. However, the surface temperature of cooked meat occurred freezing when the final pressure in the vacuum chamber was 0.3kPa. Therefore, in order to reduce the cooling time and avoid freezing, the temperature of vapour-condenser should be set around 30 ~-40 and the final pressure in the vacuum chamber can be defined at from 0.4kPa to 0.61kPa.
℃ ℃
℃
℃
℃
℃ ℃
Keywords: Temperature; Pressure; Vapour-condenser; Vacuum cooling.
1 Introduction Vacuum cooling is a rapid evaporative cooling method. Vacuum cooling has been successfully used to cool vegetables and flowers since the 1950s [1]. In the recent years, for the safety of foods, a rapid cooling treatment after cooking process should be used to minimize the growth of surviving organisms. Compared with the conventional cooling methods including air-blast, slow-air and water-immersion cooling, vacuum cooing has many advantages. Therefore, many researches have highlighted the applications of vacuum cooling for the cooked meats [2-4]. In addition, heat and mass transfer characteristics during vacuum cooling have been investigated [5]. *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 41–52, 2011. © IFIP International Federation for Information Processing 2011
42
T. Jin, G. Li, and C. Hu
Predictive models can provide much valuable information for the cooling process of large cooked meat joints under broad experimental conditions within a short time. Wang and Sun have developed a mathematical model for describing the vacuum cooling process of the large cooked meat joints [6-9]. Nomenclature
-cold load of vapour-condenser, W ; R -gas constant for water vapour, J ⋅ mol h -sublimation heat of ice, J ⋅ kg ; T -the Kelvin temperature, K ; Q0 v
−1
⋅ K −1
−1
;
vs
m − mass flux, kg ⋅ s −1 ;
-specific volume, m ⋅ kg ; -pressure, Pa ; -the diffusivity, m ⋅s ; Greeks ρ -density kg ⋅ m ; −1
3
v P D
2
−1
−3
λ − thermal conductivity, J ⋅ m −1 ⋅ K −1 ⋅ s −1 ;
Subscripts fr frost layer
- ; v -vapour;
ice − ice layer;
A vacuum cooler is a machine to maintain the defined vacuum pressure in a sealed chamber, where the boiling of the water in the cooked meats occurs to produce the cooling effect. Theoretically, only the speed of vacuum pump is high enough to produce the defined vacuum pressure in the vacuum chamber. However, at a low pressure, the volume ratio of steam and water is very large. For example, when the pressure is 1073 Pa, the corresponding saturation temperature is 8 , the specific 3
℃
volume is 120.851 m kg . If the entire vapour is evacuated only through the vacuum pump, the speed of vacuum pump should be very large, many vacuum pumps are required in the vacuum cooler, which is obviously unsuitable. In order to remove the large amount of water vapour and keep the cooling cycle within a reasonable length of time, the vapour-condenser is used to economically and practically handle the large volume of water vapour by condensing the vapour back to water and then draining it through the drain valve. The vacuum pump and the vapour-condenser in the vacuum cooling system are used to remove the water vapour evaporated from the cooked meats. Wang and Sun [10] analysed the effect of operating conditions of a vacuum cooler on the cooling performance for large cooked meat joints by a validated mathematical model, they concluded that the temperature of the vapour-condenser should
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber
℃,
43
be above 0 because water freezes on the outside surface of the condenser when the temperature is below 0 . It is well known that the boiling point changes as a function of saturation pressure, for a boiling temperature of 0 the saturation pressure will be 609 Pa. Therefore, in order to avoid freezing, the final pressure in the vacuum chamber is usually above 609 Pa. On the base of previous literatures, in the current study, vacuum cooling of cooked meats were conducted to analyze the effects of the final pressure in the vacuum chamber below 609 Pa and the temperature of vapour-condenser below 0 on the cooling rate of cooked meats.
℃
℃,
℃
2 Materials and Methods 2.1 Samples Preparation The raw bone-out pork used in experiments was brought from a local supermarket. Then, the samples were cooked in water through the oven (Type of the oven is RFP130Y, China) until the samples were at a uniform temperature. Then, the cooked meat was put and cooled in the vacuum chamber. 2.2 Experimental Setup A laboratory-scale vacuum cooler as shown in Fig. 1 was built by Shanghai Pudong Freezing Dryer Instruments Co. Ltd. (Shanghai, China). Vacuum cooler has four basic components: a vacuum chamber, a vacuum pump, a vapour-condenser and a refrigeration system. The volume of vacuum chamber was approximately 0.3m3. The rotary vane vacuum pump (Type 2XZ-2) with the pumping speed of 7.2m3h-1 and rotary speed 1400 rev min-1 was used to evacuate the air in the vacuum chamber and the vapour evaporated from the products from atmospheric pressure to the defined vacuum pressure. The final vacuum pressure in the vacuum chamber is regulated by the bleeding valve. The vapour-condenser is an evaporator in the refrigeration system and a condenser capturing water vapour evaporated from the cooked meats during vacuum cooling. The cooling coil of vapour-condenser is set up in a stainless cylindrical steel, which is enclosed with 30 mm thickness polyurethane foam to prevent heat transfer. The stainless cylindrical steel with vapor-condenser is defined as cold trap. 2.3 Data Collection
℃
A set of T-type copper-constantan thermocouples with an accuracy of ± 0.1 are used to record the temperature distribution of cooked meats and the temperature of the cold trap. The pressure sensor (model CPCA-130Z), a capacitance membrane gauge with an accuracy of ± 1 Pa was used to measure the vacuum pressure in the chamber. The data collection and control signals, such as pressure and temperature were conducted by I-7000, a family of network data acquisition and control modules. The control module was connected with software called “King of Combination” (Beijing Asia Control Automatic Software Co. Ltd.). In order to eliminate the error of the second conversion, the temperatures received by the computer were demarcated by a second scale standard mercury thermometer with a measurement range 0~100 .
℃
44
T. Jin, G. Li, and C. Hu
1-bleeding valve; 2-weight sensor; 3-sample; 4-thermal couple; 5-pressure sensor; 6-vacuum chamber; 7-electronic balance; 8-compute; 9-temperature controller; 10-coolant outlet; 11coolant inlet; 12-cold trap; 13-vacuum pump; 14-pressure controller; 15-I-7018P module Fig. 1. Schematic diagram of the vacuum cooler system
3 Results and Discussion 3.1 The Effect of Temperature of Vapour-Condenser on Cooling Rates during Vacuum Cooling The vapour-condenser, which is an auxiliary vacuum pump, is normally used to remove the large amount of water vapour generated by condensing the vapour back to water and draining the water out of the vacuum chamber. The effect of temperature of vapour-condenser on cooling rates is shown in Fig. 2. It can be seen from Fig. 2 that the cooling rate of cooked meat can increase with the reduction of temperature of vapour-condenser. During vacuum cooling, the temperature of vapour-condenser of below 0 was used. However, if the temperature of vapour-condenser was too low, the cooling rate of cooked meat can decrease. If 0.15 kg of cooked meat was cooled, the average temperature of cooked meat can be reduced from 61.4 to 4.7 within 30 min at the vapour-condenser temperature of -2 . In addition, when the temperature of vapour-condenser was further reduce from -2 to -39 , the total cooling time can be obviously reduced from 30 min to 20 min. On the other hand, if the mass of cooked meat was 0.5 kg, the temperature of vapour-condenser was reduced continuously to –71 , the average temperature of cooked meat decreased only from 74 to 26.1 within 42 min. In the same mass of cooked meat, the average temperature of cooked meat can be reduced from 60.1 to 5.7 within 50 min at the vapour-condenser temperature of -35 . During vacuum cooling, the mass of cooked
℃
℃
℃
℃
℃
℃
℃
℃ ℃
℃
℃
℃
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber
45
meat has also an effect on the cooling rate, which is accorded with Wang and Sun’s [10] experimental result. Wang and Sun [10] think that the temperature of vapourcondenser should be set at around 2.5 above 0 in order to avoid freezing of water on the outside surface of the cold trap. However, author thinks that the temperature of vapour-condenser should be set below 0 . Because when the temperature of vapour-condenser is below 0 , the water vapour become frost through solidify on the surface of the vapour-condenser, water vapour can be easily trapped in the vapourcondenser. It can be found that the different temperatures of vapour-condenser below 0 have an effect on the cooling rate of cooked meat. However, if the temperature of vapour-condenser is too low, the cooling time can be increase adversely, which can be expressed by the formation of the frost on the surface of cold trap.
℃
℃
℃
℃
℃
Fig. 2. Effect of temperature of vapour condenser on cooling rate during vacuum cooling
3.2 The Formation Mechanism of the Frost on the Surface of Vapour-Condenser Fig. 3 shows the formation process of frost. The sensible heat is transferred from the water vapour in the vapour-condenser to the frost surface by the temperature difference driving force between the water vapour and the frost surface. Some of the transferred moisture deposits on the frost layer, causing the frost layer to grow. The remainder diffuses into the frost layer. The heat of sublimation caused by the phase change of the added frost layer is transferred through the frost layer. The latent heat
46
T. Jin, G. Li, and C. Hu
and sensible heat transferred from the water vapour are then transferred through the frost layer by conduction. The water vapour diffusing into the frost layer changes phase within the frost layer. The frost density increases as a result of this process. The frost layer is a porous medium composed of ice crystal and air. The ice crystal has different shapes during the formation of the frost layer. Ice crystal shapes are classified into main forms: plate-like forms and column-like forms. The microscopic structure of ice crystal is shown as in Fig.4. Sensible heat transfer Latent heat transfer Frost surface Heat transfer by conduction
Phase change
Frost layer
Water vapor diffusion
Vapor- condenser surface
Fig. 3. The formation process of the frost
Fig. 4. Ice crystal shape (1) Plate-like forms: (a) plate, (b) simple sectored plate, (c) dendritic sectored plate, (d) fern-like stellar dendrite; (2) Column-like forms: (e) needle crystal, (f) hollow column, or sheath-like crystal [11]
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber
47
During the formation of the frost, the mass flux through water vapour diffusing into the frost layer can be calculated by the Clapeyron-Clausius equation. The expression is as follows [12]:
m fr =
Q0 ⎡ ⎛ ρ f ⎞ 0.5 ⎤ λ fr RT (vv − vice ) ⎢1 + ⎜⎜ r ⎟⎟ ⎥ ⎢⎣ ⎝ ρ ice ⎠ ⎥⎦ hvs + ⎛ ρf ⎞ Dv [hvs − Pv (vv − vice )]⎜⎜1 − r ⎟⎟ ⎝ ρ ice ⎠ 2 fr
Where
(1)
Q0 is the refrigeration load of vapour-condenser;
hvs is the sublimation heat of ice; R is the gas constant; T fr is the surface temperature of the frost layer; vv and vice are respectively specific volume of water vapour and ice;
ρf r
and
ρ ice
are respectively density of frost layer and ice;
Pv is the partial pressure of water vapour; Dv is the diffusivity of water vapour;
λ fr
is the thermal conductivity of the frost layer, the expression is as follows [13]:
λ fr = 0.02422 + 7.214 × 10 −4 ρ fr + 1.1797 ×10 −6 ρ fr 2
℃
(2)
The temperature of vapour-condenser is below 0 , The water vapour evaporated from the cooked meats will become the frost at the surface of vapour-condenser. Fig. 5 shows that the comparison of the cold trap between before and after vacuum cooling of cooked meat. It can be obviously found that the frost occurs at the surface of cold trap after vacuum cooling of cooked meat. The latent heat of sublimation is released in the cold trap, which can increase the temperature of vapour-condenser. The variation of temperature of vapour-condenser is shown in Fig. 6. It can be seen from Figs. 2 and 6 that when the initial temperatures of cold trap were at -2 , -39 and -35 , respectively, the temperatures of cooked meat were reduced to about 5 , the temperature of vapour-condenser was reduced from -2 , -39 and -35 to –70.4 , -70.4 and –71.9 , respectively during vacuum cooling. It can be found that the temperature of vapour-condenser can be reduced continuously during vacuum cooling. Because the condensation ability of vapourcondenser is not smaller than the required one, the condenser can efficiently condense all the generated water vapour during the cooling. However, when the initial temperature of cold trap reached at -71 , the temperature of cold trap had no variation during
℃ ℃
℃ ℃ ℃ ℃ ℃ ℃ ℃ ℃
℃
48
T. Jin, G. Li, and C. Hu
a. Cold trap before vacuum cooling of cooked meat
b. Cold trap after vacuum cooling of cooked meat Fig. 5. Comparison the cold trap before vacuum cooling of cooked meat with after vacuum cooling of cooked meat
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber
49
Fig. 6. The variation of temperature of vapour condenser in different experimental conditions
Fig. 7. Effect of the pressure in the vacuum chamber on the surface temperature of cooked meat during vacuum cooling
50
T. Jin, G. Li, and C. Hu
vacuum cooling, which can be seen from Fig. 6. At the same time, the temperature of cooked meat reduced only from 74 to 26.1 within 42 min, which is because the frost and ice at the surface of cold trap result in a significant heat resistance.
℃
℃
3.3 The Effect of the Pressure in the Vacuum Chamber on the Cooling Rate during Vacuum Cooling Fig. 7 gives the effect of the pressure in the vacuum chamber on the surface temperature of cooked meat during vacuum cooling. Four different final pressure, 0.3kPa, 0.4kPa, 0.5kPa and 0.61kPa, in the vacuum chamber were chosen during vacuum cooling. It can be seen from Fig. 7 that if the final vacuum pressure in the vacuum chamber is 0.61kPa, the surface temperature of cooked meat can be reduced from 50 to 5.2 within 40 min. When the final vacuum pressure in the vacuum chamber was reduced from 0.61kPa to 0.5kPa, the surface temperature of cooked meat can be reduced from 43.5 to 4 within 31 min. The final vacuum pressure in the vacuum chamber was kept at 0.4kPa, the surface temperature of cooked meat varied from 51.2 to 5.6 within 18 min. In the same case, the mass of cooked meat is 0.3kg, the final vacuum pressure in the vacuum chamber varied from 0.4kPa to 0.61kPa, the cooling time will increase from 18 min to 40 min. If the mass of cooked meat increased from 0.3kg to 0.6kg, the final vacuum pressure in the vacuum chamber is further reduced from 0.4kPa to 0.3kPa, the surface temperature of cooked meat can be reduced from 41.6 to 2.7 within 25 min. On the other hand, during vacuum cooling, it can be found when the final vacuum pressure in the vacuum chamber was 0.3kPa, the minimum surface temperature of cooked meat was –0.5 , which shows that water freezes on the surface of cooked meat and has a negative effect on the cooked meat. It is well known that the boiling point changes as a function of saturation pressure, for a boiling temperature of 0 the saturation pressure will be 609Pa. However, the experimental results show that the surface temperature of cooked meat is above 0 , when the vacuum pressure in the vacuum chamber is between 0.4kPa and 0.6kPa. If the vacuum pressure in the vacuum chamber is further reduced to 0.3kPa, the surface temperature is below 0 . This means that the vacuum pressure in the vacuum chamber can be reduced to below 0.6kPa. At the same time, it should be noted that the vacuum pressure in the vacuum chamber should be above 0.4kPa.
℃
℃
℃
℃
℃ ℃
℃
℃
℃
℃,
℃
℃
4 Conclusion
℃ ℃ ℃ ℃
The temperature of vapour-condenser below 0 and the final pressure in the vacuum chamber below 0.61kPa during vacuum cooling were experimentally analysed. The temperature of vapour-condenser, -2 , -35 , -39 and -71 , and the final pressure in the vacuum chamber, 0.3kPa, 0.4kPa, 0.5kPa and 0.61kPa, were chosen during vacuum cooling. The experimental results showed that water vapour becomes the frost on the surface of vapour-condenser when the initial temperature of vapourcondenser is below 0 , which is helpful to trap water vapour for vapour-condenser. However, if the temperature of vapour-condenser is reduced continuously to -71 ,
℃
℃
℃
Influences of Temperature of Vapour-Condenser and Pressure in the Vacuum Chamber
51
the cooling rate will not increase. Therefore, the temperature of vapour-condenser should be set around -30 ~-40 . At the same time, it can be also found that the cooling time for vacuum cooling can be reduced when the final pressure in the vacuum chamber varied from 0.4kPa to 0.61kPa. In addition, the surface temperature of cooked meat was above 0 . However, the surface temperature of cooked meat occurred freezing when the final pressure in the vacuum chamber was 0.3kPa. It can be suggested that the final pressure in the vacuum chamber can be set between 0.4kPa and 0.61kPa. In a word, it is feasible that the temperature of vapour-condenser below 0 and the final pressure in the vacuum chamber below 0.61kPa. In order to reduce the cooling time, the temperature of vapour-condenser should be set around -30 ~40 and the final pressure in the vacuum chamber can be defined at from 0.4kPa to 0.61kPa.
℃ ℃ ℃
℃ ℃
℃
Acknowledgements Funding for this research was provided by Henan Provincial Department of Education (P. R. China).
References 1. Briley, G.C.: Vacuum cooling of vegetables and flowers. ASHRAE Journal 46(4), 52–53 (2004) 2. McDonald, K., Sun, D.-W., Kenny, T.: The effect of injection level on the quality of a rapid vacuum cooled cooked beef product. Journal of Food Engineering 47, 139–147 (2001) 3. Burfoot, D., Self, K.P., Hudson, W.R., Wilkins, T.J., James, S.J.: Effect of cooking and cooling method on the processing times, mass losses and bacterial condition of large meat joints. International Journal of Food Science and Technology 25, 657–667 (1990) 4. Desmond, E.M., Kenny, T.A., Ward, P., Sun, D.-W.: Effect of rapid and conventional cooling methods on the quality of cooked ham joints. Meat Science 56, 271–277 (2000) 5. Sun, D.-W., Wang, L.J.: Heat transfer characteristics of cooked meats using different cooling methods. International Journal of Refrigeration 23, 508–516 (2000) 6. Wang, L., Sun, D.-W.: Modelling vacuum cooling process of cooked meat—part 1: analysis of vacuum cooling system. International Journal of Refrigeration 25, 854–861 (2002) 7. Wang, L., Sun, D.-W.: Modelling vacuum cooling process of cooked meat—part 2: mass and heat transfer of cooked meat under vacuum pressure. International Journal of Refrigeration 25, 862–871 (2002) 8. Sun, D.-W., Hu, Z.: CFD predicting the effects of various parameters on core temperature and weight loss profiles of cooked meat during vacuum cooling. Computers and Electronics in Agriculture 34, 111–127 (2002) 9. Sun, D.-W., Hu, Z.: CFD simulation of coupled heat and mass transfer through porous foods during vacuum cooling process. International Journal of Refrigeration 26, 19–27 (2003)
52
T. Jin, G. Li, and C. Hu
10. Wang, L., Sun, D.-W.: Effect of operating conditions of a vacuum cooler on cooling performance for large cooked meat joints. Journal of Food Engineering 61, 231–234 (2004) 11. Na, B., Webb, R.L.: New model for frost growth rate. International Journal of Heat and Mass Transfer 47, 925–936 (2004) 12. Kondepudi, S.N., O’Neal, D.L.: Performance of finned tube heat exchangers under frosting conditions. International Journal of Refrigeration 16(3), 175–180 (1993) 13. Yonko, J.D., Sepsy, C.F.: An investigation of the thermal conductivity of frost while forming on a flat horizontal plate. ASHRAE Trans. 73(2), 111–117 (1967)
Inspection of Lettuce Water Stress Based on Multi-sensor Information Fusion Technology Hongyan Gao, Hanping Mao, and Xiaodong Zhang Key Laboratory of Modern Agricultural Equipment and Technology, Ministry of Education & Jiangsu Province, Jiangsu University, Zhenjiang 212013, China
[email protected] Abstract. Characteristics of reflection spectrum, multi-spectral images and temperature of lettuce canopy were gained to judge the lettuce’s water stress condition which could lead to a precise, rapid & stable test of lettuce moisture and enlarged the models’ universality. By the extraction of lettuce’s multi-sensor characteristics in 4 different levels, quantitative analysis model of spectrum including 4 characteristic wavelengths, characteristic model of multispectral image and CWSI were established. These multi-sensor characteristics were fused by using the BP artificial neural network. Based on the fused multisensor characteristics, the lettuce moisture evaluation model was established. The results showed that the correlation coefficient of multi-spectral images model, spectral characteristics model and information fusion model were in turn increased, the correlation coefficients were respectively 0.8042 0.8547 and 0.9337. It was feasible to diagnose lettuce water content by using multi-sensor information fusion of reflectance spectroscopy, multi-spectral images and canopy temperature. The correct rate and robustness of the discriminating model from multi-sensor information fusion were better than those of the model from the single-sensor information.
、
Keywords: Lettuce, Water stress, Information fusion.
1 Introduction Leaf stomata conductance, leaf water potential, transpiration rate, plant stem diameter changes and soil moisture, etc. indirectly illustrated the crop water stress and water requirement. Or the plant water stress is determined using dry weight method by collecting in vivo samples. The traditional testing methods have the shortages of low precision and affecting crop growth ,and it is not conducive to the promotion application, because in sampling and data analysis are time-consuming. Modern diagnostic methods are mainly spectroscopy, visual images and canopy temperature. However, canopy reflectance spectrum and images have interactions as nitrogen, water and leaf area index, spectrum and image are affected by the crop canopy structure, environmental factors and others. Therefore, a single detection method can not accurately and comprehensively explain the water stress. The discussion group proposes inspection of lettuce water stress based on the multi-sensor information fusion technology. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 53–60, 2011. © IFIP International Federation for Information Processing 2011
54
H. Gao, H. Mao, and X. Zhang
Multi-sensor information fusion might comprehensively process multi-source information which comes from some different sensors, so it can obtain more accurate and reliable conclusions [1]. Multi-sensor information fusion can greatly avoid the limitations of a single sensor and improves the performance of the system [2]. Because the sensors provide some uncertain information, multi-sensor information fusion technology is essentially a non-deterministic reasoning and decision-making process [3]. Multi-sensor information fusion can be divided into three different layers which were the decision layer fusion, feature layer fusion and the raw data layer fusion [4]. The research is the optimization and combination of spectra, images, and canopy temperature, because the spectrum analyzer, machine vision systems and other equipments are abstracted into different types of sensors, the features are different and collect different physical quantities, the information pattern and span are comparatively large, environment and objectives with time-varying features, crop characteristics are complex, so it does not suitable for the decision layer fusion and the raw data layer fusion. More practical option is feature layer fusion, the fusion not only retain a sufficient number of original information, but also achieve a level of data compression, contributes to real-time processing[5].The concept of feature layer fusion is different features gather to form the new summary feature set, and then making decisions accordingly[6]. The research obtains lettuce canopy spectrum, image information, canopy temperature and environmental temperature and humidity, etc. Then establish spectra model, mage model and the water stress index model. Ultimately, used BP neural network training samples and verification, water stress conditions on the lettuce is rapidly and non-destructively inspected.
2 Experimental Design and Sample Training 2.1 Instrument and Equipment Spectrum measuring equipment is the United States ASD FieldSpec®3 handheld portable spectrum analyzer, the range 350~2500nm; at 350~1000nm, sampling interval is 1.4nm, resolution is 3nm; in 1000 ~ 2500nm, sampling interval is 2nm, resolution is 10nm. High-precision analytical balance weighs the quality of whole lettuce, accuracy is 0.1mg. Canopy multi-spectral image utilizes MS-3100 multi-spectral digital progressive scan camera, imaging spectral range is 400~1100nm, resolution is 1039×1392. Canopy temperature utilizes the TI50 infrared thermal imaging instrument, the range is -20~305 , accuracy is 0.07 .
℃
℃
2.2 Samples Training Experiment started in Jiangsu university modern agricultural equipment and technology Venlo-type greenhouse. The variety is Italian anti-bolting lettuce. According to the Yamazaki Nutrient solution, the samples were divided into four levels, each level had 12 lettuce, so the predict set had 24 samples, calibration set had 20 samples. Four levels were: Group 1(W1) ensure adequate water were supplied throughout the
Inspection of Lettuce Water Stress Based on Multi-sensor Information Fusion Technology
55
growing season; Group 2 (W2), 3 (W3), 4 (W4) irrigated the standard formula of 75% 50%, 25% concentrate. 2.3 Experiment Design Lettuce samples (four levels) Spectral model
Multi-spectral image model
CWSI
Feature level fusion of based on BP neural network
Diagnostic assessment model of lettuce water Fig. 1. The flow chart of lettuce water stress inspection based on multi-sensor information fusion technology
3 Results and Analysis 3.1 The Quantitative Analysis of Lettuce Moisture Content Based on Spectrum Technology Fig. 2 showed the lettuce canopy reflectance spectra in the different water stress. Combined with previous studies of discussion group, and referenced to the USDA researchers came to the main biochemistry components of the spectral absorption characteristics [7], wavelength sensitive of water-related mostly concentrated in the near infrared band. As Fig. 2 shown, spectral reflectance of lettuce had significant difference in different water stress at the water sensitive bands. 0.7
Spectral reflectance
0.6 0.5 0.4 0.3 0.2 0.1 0 350
850
1350
1850
2350
Wavelength/nm
Fig. 2. Lettuce canopy reflectance spectra under different water stress
56
H. Gao, H. Mao, and X. Zhang
In order to eliminate offset and drift which caused the spectral bias, highlight the hidden information and identification of samples. The first order derivative was carried out on the spectrum and conducived to extract characteristic wavelength. In first, all spectral points were divided into 4-sensitive bands: 1220-1300nm ,14101490nm, 1620-1700nm ,1900-1970nm, then removed the wavelength by stepwise regression[8] and got the sensitive wavelength of related to the lettuce canopy water stress: 1267nm 1443nm 1661nm 1921nm. So as to eliminate the impact of multicollinearity, so the four wavelengths for partial least squares regression analysis[9], When two principal component score were extracted, cumulative contribution rate greater than 0.85. Ultimately obtained PLS regression model based on four sensitive wavelengths (Xi):
、
、
、
y = 4670.62 − 3786.43X1 − 2994.85X 2 − 2990.34X 3 + 943.02X 4
(1)
Then 20 samples of spectral data tested the model, the correlation coefficient between dry water content of lettuce canopy measured and predictive value was 0.8547.
Fig. 3. Lettuce canopy reflectance spectra under different water stress
3.2 The Quantitative Analysis of Lettuce Moisture Content Based on Multi-spectral Imaging Technology The six channels lettuce canopy images were simultaneously acquired by the MS3100 multi-spectral digital progressive scan camera, they were R G B RGB IR and CIR. This method not only contributed to extract image features of all the independent channels, but also easily achieved multi-spectral image pixel-level operation and integration(Image registration was not required). Image processing based on MATLAB software. Median filtering method of 3 × 3 window would eliminated isolated noise points, reduced the image blurring, so it was used for image preprocessing. And two-dimensional maximum entropy segmentation was used for background segmentation; this method preserved more image information of crop canopy and contributed to the image feature extraction. Experiment used gray feature extraction method, finally AIR810 and AIR940 (AIRk is near infrared
、、、
、
Inspection of Lettuce Water Stress Based on Multi-sensor Information Fusion Technology
57
spectroscopy 810nm and 940nm canopy image mean gray value) significantly correlated to lettuce canopy water content. Establishing lettuce canopy water content forecast model by SPSS 13.0 for multiple linear regression analysis, including AIR810 and AIR940 image features variable: y = 24 .764 + 124 .729 AIR 940 + 63 .775 AIR 810
(2)
Then 20 samples of spectral data tested the model, the correlation coefficient between dry water content of lettuce canopy measured and predictive value was 0.8042. 3.3 Canopy Water Stress Index (CWSI) Model Establishment
The TI50 infrared thermal imaging instrument obtained lettuce canopy temperature and real-timely monitored environmental temperature and humidity. According to the CWSI empirical model by Idso in the literature [10], as follows:
CWSI =
(Tc − Ta ) − (Tc − Ta )Π (Tc − Ta )ul − (Tc − Ta )Π
(3)
(Tc − Ta )Π = A + B × VPD V P D = 0 .6 1 1 × e
1 7 .2 7 × T a T a + 2 3 7 .3
(4)
× (1 −
(5)
RH ) 100
(Tc − Ta ) ul = A + B × V PG
(6)
In the formula: Tc-the crop canopy temperature, °C; Ta-air temperature, °C; (Tc-Ta)Π-lower limit of the difference temperature between canopy and air, °C; (Tc-Ta)ul-limit of the difference temperature between canopy and air, °C; VPD-air vapor pressure deficit, hPa. A, B-experience factor; VPG-the difference of air saturated vapor pressure between when the air temperature were Ta and Ta+ A, hPa. y = -1.4647x + 4.8287 R2 = 0.8969
4 3 2
C 1 ° 0 / a 1 T -1 c T -2 -3
2
3
4
5
-4 -5
VPD/Kpa Fig. 4. The relationship of (Tc-Ta) and VDP
6
58
H. Gao, H. Mao, and X. Zhang
The model of difference temperature between canopy and air by data analysis:
T c − T a = 4 . 8287 − 1 . 4647 VPD
(7)
In the condition of full water supply, when VPD was 5.86, the difference temperature between canopy and air was minimum, it regarded as lower limit of CWSI. So (Tc-Ta)Π was -3.7544. When lettuce was severe water stress, canopy temperature reached the maximum during the experiment, it regarded as upper limit. So (Tc-Ta)ul was 1.3241( ). Lettuce CWSI model:
℃
CWSI =
(8)
Tc - Ta + 3 .7544 5 .0785
3.4 The Model Based on BP Neural Network Information Fusion
BP neural network has strong fault tolerance, distributed, storage, self-learning, adaptive, self-organization, nonlinear dynamic capabilities and handle complex environments[11].Therefore, using better self-learning and adaptive capacity, lettuce water stress condition was evaluated by BP neural network. Table 1. Samples predicted values and the measured values
Sample Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Average
Predictive Value (%) 2984.269 2712.595 2440.624 2402.126 2290.337 1934.597 1852.465 1839.795 1791.997 1780.012 1710.706 1661.028 1690.231 1577.302 1603.096 1687.403 1575.115 1436.798 1206.933 1081.009
Measured Value (%) 2812.366 2508.462 2604.845 2598.000 2499.381 2161.101 2093.596 1721.290 1623.837 1968.000 1603.889 1798.400 1864.688 1704.605 1428.125 1490.943 1449.184 1625.636 1415.647 1174.524
Relatively Error (%) 6.112 8.138 6.304 7.539 8.364 10.481 11.518 6.885 10.356 9.552 6.660 7.639 9.356 7.468 12.252 13.177 8.690 11.616 14.743 7.962 9.240
Inspection of Lettuce Water Stress Based on Multi-sensor Information Fusion Technology
59
The research used a 3-layer structure of BP neural network for feature level data fusion. There were lettuce canopy reflectance spectra characteristic wavelength 1267nm, 1443nm, 1661nm, 1921nm and the image feature parameter AIR810, AIR940, CWSI as the input, dry basis moisture content measured as the output. Error index and the training step were respectively sited to 0.001 and 0.05, hidden nodes was 10. Using BP neural network to predict the same test set (Table 1). The average relative error of predicted and measured values was 9.24%, correlation coefficient R was 0.9337.
4 Conclusion Object of study choose four different moisture content of lettuce, lettuce canopy water stress was evaluated by spectral characteristics, image feature information, canopy temperature and Environmental temperature and humidity. Spectral analysis model and the image model were established, then verifying the model, the correlation coefficient between water content of lettuce canopy measured and predictive value were 0.8547 and 0.8042. The results showed that lettuce canopy water stress evaluation method based on the spectrum, multi-spectral image and the CWSI of multi-sensor information fusion technology was feasible, and the correlation coefficient was 0.9337. Model of accuracy and stability were higher than a single information model. Results of the research for multi-sensor information fusion technology could regarded as reference to the rapid and accurate detection of lettuce water.
Acknowledgements This work was supported by a grant from the National High Technology Research and Development Program of China (863 Program)(No.2008AA10Z204 and 2008AA102208), "333 Talent Project" in Jiangsu Province.
References 1. He, Y., Wang, G., Lu, D., Peng, N.: Multi-sensor data fusion and applications. Electronic Industry Press, Beijing (2007) 2. Xiao, W., Li, X., Li, P., Feng, Y., Wang, W., Zhang, J.: Near infrared spectroscopy and machine vision information fusion soil moisture detection. Transactions of the CSAE 25, 14–17 (2009) 3. Ma, G., Zhao, L., Li, P.: Based on Dempster Shafer evidence of multi-sensor information fusion technology and its application. J. Modern Electronic Technology 19, 41–44 (2009) 4. Yan, H., Huang, X., Wang, M.: Multi-sensor data fusion technology and its application. J. Sensor Technology 24, 1–4 (2005) 5. Chen, Q., Zhao, J., Cai, J.: Based on near infrared spectroscopy and machine vision information fusion technology of multi-judge the quality of tea. J. Transactions of the CSAE 24, 5–10 (2008) 6. Wang, R.: Information fusion. Science Press (2007)
60
H. Gao, H. Mao, and X. Zhang
7. Pu, R., Gong, P.: Hyper spectral remote sensing and its application. Higher Education Press, Beijing (2000) 8. Zheng, Y., Zhang, J., Chen, X., Shen, X., Zhang, T.: Based on stepwise regression of nearinfrared spectral information extraction and model. J. Spectroscopy and Spectral Analysis 24, 675–678 (2004) 9. Wang, H.: Partial least squares regression method and its application. National Defense Industry Press, Beijing (1999) 10. Cui, X., Xu, L., Yaun, G., Wang, W., Luo, Y.: Based on the temperature of the summer maize canopy water stress index model study. J. Transactions of the CSAE 21, 22–24 (2005) 11. Xiong, Y., Wen, Z., Wang, M.y.: Based on neural network spectral recognition system design and analysis. J. Spectroscopy and Spectral Analysis 27, 139–142 (2007)
Measurement of Chili Pepper Plants Size Based on Mathematical Morphology Yun Gao, Xiaoyu Li, Kun Qi, and Hong Chen College of Engineering, Huazhong Agricultural University Wuhan, China
[email protected] Abstract. Since chili pepper plant size directly reflects the state of plant growth, a method for pepper measurement of plants size was discussed here. Pepper plants were shot from above once per week in the greenhouse since being field planted in spring. The method of processing the pepper plant images was studied, in which the image segmentation of combination of color space and the image morphological operations were applied. And the major axis and minor axis of pepper plant, for describing the size of the plant, were calculated from single connected component in the image being processed. According to the method, a program for pepper plant size measurement based on MATLAB was developed. Experimental results have demonstrated that the method is more reasonable and accurate than artificial measure. Keywords: pepper plant; size measurement; segmentation; morphological operation; major axis and minor axis.
1 Introduction Chili pepper, which plays an important role in the year-round vegetable supply in china, is an important commercial-orientated crop in the country[1,2]. During the cultivation of chili pepper plants, the growth state and morphological directly influences the suitability of a plant for cultivation, its overall yield and its economic coefficient[3]. The time of each growth phase, the number of leaves, weight of fresh leaf, leaf area, thickness of leaf, size of leaf, and so on are used to describe the growth state. However, the size of plants, as the intuitive and important factor to describe the growth stage and growth state, has less been studied, because of the difficult measurement. As comparing the differences in size between the same capsicum species does help research on capsicum cultivation techniques and improve the yield and quality of pepper. In this work, we developed a method to detect the size of capsicum plants using computer vision technology. The chili pepper plants were photographed in the greenhouse for the size measurement method developed. An algorithm, using image segmentation method to separate pepper plant from the background, and image binarization method to make the image black and white, after that, the morphology method was utilized to make single frame pepper plant image into a single connected graph. Finally, the longest diameter D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 61–70, 2011. © IFIP International Federation for Information Processing 2011
62
Y. Gao et al.
and the shortest diameter, as plant morphology parameters, of the single connected graph, were introduced. Experiments verified that the algorithm was effective, with comparing measurement data with the tape to data calculated by the algorithm.
2 Image Acquisition 50 chili pepper plants, which were planted in the spring in the greenhouse of Hubei Academy of Agricultural Science and Technology, photos have been taken for study, by using single μ300 Olympus digital camera and a tripod metal photographic PTZ. Chili pepper seedlings were transplanted from the seedbed to the greenhouses, as planting spacing of 40cm and seedling spacing of 45cm. One week later, chili pepper plants were photographed once a week for seven weeks. 150 pictures were collected each time, and three pictures were taken from one chili pepper plant. In the photo collection, the camera was placed on the tripod metal photographic PTZ, just perpendicular to the plant and shot the plant from above, as shown in Figure 1, in which H is the distance from the camera lens to the ground. Between three times shooting, the camera was rotated 120 degrees in the horizontal direction. The image resolution is 1024×768, each image is saved as a JPEG file. To improve the adaptability of the analysis method, the shooting was not under extra lighting but natural light. The first three weeks after the beginning of image acquisition, each image contained only one plant. From the fourth week, chili pepper plants grew staggered, and not suitable for image acquisition. So the study object in this paper is the images acquired from the shooting of the first three weeks. Fig. 1 shows how the pictures have been taken, in which H is the vertical distance from the actual shooting of the camera lens to the ground pepper cultivation.
,
Fig. 1. Sketch map for shooting method
3 Image Segmentation 3.1 RGB Color Image Segmentation To detect the size of chili pepper plants, pepper images need to be segmented from background. From pepper picture in Fig. 2, we can see the background of pepper
Measurement of Chili Pepper Plants Size Based on Mathematical Morphology
63
plants mainly compos of the soil covered with plastic film, the black section on the upper right corner of the image is irrigation tube road under the plastic film. As the main color of pepper plants is green, some parts of the leaves and petioles are yellow and the color is very close to the background color of the soil and the film, which makes the image segmentation very difficult. At present, there are two main methods to use the color characteristics for the color image segmentation: one is changes the two-dimensional color images into grayscale images, and grayscale threshold segmentation algorithm is used for gray image segmentation; another is based on color segmentation, and in the color space it directly limits each RGB value of the color space and separates the chili pepper plant and the background[4,5].
Fig. 2. Image of chili pepper plant
Major color spaces, used in color region segmentation today, are RGB color space and HIS color space. The images shown in Fig. 1, is segmented for the Euclidean distance[6]. The Euclidean distance between z and m is given by D (z, m ) = z − m = [(z − m)T (z − m )]
1
2
= [( z R − mR ) 2 + ( zG − mG ) 2 + ( z B − mB ) 2 ]
1
2
(1)
Wherein m stands for the RGB column vector of average color from the region of chili pepper plant to be segmented and z stands for an arbitrary point in RGB space.
⋅ is the norm of the argument, and subscripts R,G and B, stands for the RGB values of vectors m and z. Figs.3 (a)through (d) show the segmentation results with T =25, 45 ,60 and m = [96.0202 126.0374 45.5014]'. Here m is a vector of mean RGB values in the plant region. In Figs.3 (a)through (d) show when T is too small , the deterioration of plant appears in (a). when T is too large, the background cannot be segmented well from the image . To directly set the threshold of RGB values with R(i,l)= 59&B(i,l)=3&G(i,l)=92 can not have an well segmentation result, which shows in Fig. 4. In the study we found the yellow soil could be segmented well in HIS color space by using the threshold algorithm, but the plastic film and irrigation tube road couldn’t. The image processing result was shown in Figure 5 with H (i, l) 0.15 & I (i, l) #I D
:
:
." :
,"
"
:
: 1:
' "2
1:
8-1*46 >5 >4 >3 >0.3 >0.25 >0.20 >0.15 >250 >200 >150 >120 >30 >25 >20 >15
6~5 5~4 4~3 3~2.5 0.3~0.25 0.25~0.20 0.20~0.15 0.15~0.125 250~200 200~150 150~120 120~90 30~25 25~20 20~15 15~10
5~4 4~3 3~2 2.5~1.5 0.25~0.20 0.20~0.15 0.15~0.10 0.125~0.075 200~150 150~120 120~75 90~60 25~20 20~10 15~10 10~5
4~3 3~2 2~1 1.5~0.6 0.20~0.15 0.15~0.10 0.10~0.05 0.075~0.03 150~120 120~75 75~30 60~30 25~20 10~5 10~5 5~3
3~2 0.79) to better evaluate the growth state of rice mainly within near-infrared, visible light and red edge 707nm range. Above studies verified the estimation of chlorophyll and LAI through remote technique, mainly for wheat and maize, but seldom for tobacco. Tobacco is harvested through leaf, and it is significant to monitor its LAI and chlorophyll content. This study tested tobacco under different fertilization conditions, defined the change law of chlorophyll content/LAI/canopy spectral characteristics, determined the correlation of vegetation index with chlorophyll content and LAI at different waveband combinations, confirmed whether canopy spectrum could be used to estimate LAI and chlorophyll content, and provided technological means for quick harmless large-area obtaining of information on the growth state of tobacco.
2 Materials and Methods 2.1 Experiment Design 5 levels of nitrogen fertilizer: N0 (0kg/hm2), N1 (22.5kg/hm2), N2 (45kg/hm2), N3 (67.5kg/hm2) and N4 (90kg/hm2). 3 levels of organics: Low (225 kg/hm2), middle (375kg/hm2) and high (525kg/hm2). No base fertilizer, repeat each treatment for 3 times, and randomly divide into 45 blocks (12m×6m). Qin tobacco #96, at a row spacing of 1.2m and a plant spacing of 0.6m. Experiment 1: At National Modern Tobacco Demonstrative Area in Mianchi County, Sanmenxia City, Henan, China (north latitude 34º37´52.96″ and east longitude 111º44´28.24″). Surface layer 0~20cm, soil organics 12.3g/kg, total nitrogen 0.74mg/kg, effective phosphorus 9.7mg/kg, quick-acting potassium 120mg/kg, and soil pH8.2. Experiment 2: At Xuedian Village, Zhuyang Town, Lingbao City, Henan, China (north latitude 34º17´13.56″ and east longitude 110º44´29.04″). Surface layer 0~20cm, soil organics 11.9g/kg, total nitrogen 0.8mg/kg, effective phosphorus 7.2mg/kg, quick-acting potassium 132mg/kg, and soil pH7.9.
652
H. Qiao et al.
2.2 Spectral Data Collection This study made spectrometry at characteristic growth stage of tobacco: Seedling stage (June 20, 2009), resettling growth stage (July 19, 2009), vigorous growth stage (August 11, 2009), early-maturity stage (September 5, 2009) and maturity stage (September 23, 2009). Canopy spectrometer adopted portable hyperspectral radiometer of American ASD Co. (HandHeld), spectral range 325~1050nm, spectral resolution 1.4nm, field angle 25º. Make spectrometry in 3 tobacco plants of same growth state at each treatment, determine canopy spectral reflectance at 10:00~14:00 in sunny windless climate, with sensor probe vertically downwards and at 60cm vertically away from canopy top. Calibrate standard white plate before each determination, record data in 20 groups, and regard their mean as spectral reflectance value of this treatment. 2.3 Determination of Leaf Area and Chlorophyll Content LAI=Length×width×0.6345 [10] Chlorophyll content: SPAD-502 chlorophyll detector of Japan MINOLTA Co. was used to determine spectrum and chlorophyll content in tobacco, i.e. Specialty Products Agricultural Division (SPAD). According to the absorptivity of leaf chlorophyll to colored light, SPAD-502 chlorophyll detector determined the chlorophyll content in leaf by determining the intensity of emission light at a certain wavelength and that of light through leaf. SPAD-502 chlorophyll detector had 2 emission light sources to emit red light of maximum 650nm and infrared light of maximum 940nm respectively. Chlorophyll absorbed red light of 650nm, but did not absorb infrared light of 940nm which was emitted and received to mainly eliminate the influence of leaf thickness on determination results. After reaching leaf, some of red light was absorbed by leaf chlorophyll, and the remaining passed through leaf and was converted into electric signal through receiver. The chlorophyll content in leaf was determined by comparing the intensity of emission light with that of light received by receiver (SPAD unit): SPAD = Klg
⎡ IRt / IR 0 ⎤ ⎢⎣ Rt / R 0 ⎥⎦
Wherein, K: Constant; IRt: Intensity of received 940nm infrared light; IR0: Intensity of emission infrared light; Rt: Intensity of received 650nm red light; and R0: Intensity of emission red light. 2.4 Data Analysis This study analyzed the correlation of several spectral parameters with chlorophyll content and LAI in tobacco canopy. This study selected significantly-related sensitive waveband and spectral parameters, established chlorophyll content and LAI monitoring model through regression analysis, and optimized the equation through estimated standard error (SE) and determination coefficient (R2). Spectral parameters were shown in Table 1.
Study on Relationship between Tobacco Canopy Spectra and LAI
653
Table 1. Hyperspectral parameters used in this study Spectral parameter
Algorithm formula and definition RVI=NIR/RED NDVI=(NIR-RED)/(NIR+RED) NDVIgreen=(NIR-Green)/(NIR+Green) SAVI=1.5*(NIR-RED)/(NIR+RED+0.5) OSAVI=(NIR-RED)/(NIR+RED+0.16)
RVI NDVI NDVIgreen SAVI OSAVI MSAVI
MSAVI = (1/2)[2 * NIR + 1 - (2 * NIR + 1)
2
- 8 * (NIR - RED)] MCARI=[(R700-R670)-0.2*(R700-R500)]*(R700/670)
MCARI
3 Results and Analysis 3.1 Raw Spectral Characteristics of Tobacco Canopy Tobacco grew slowly, rapidly and then slowly at early, middle and late growth stage respectively. For example, the leaf area and plant height of tobacco increased slowly at 60 days after the transplantation respectively, i.e. overall like S-shaped growth curve. At different growth stages, spectral curve of tobacco canopy changed at both same and different tendency, i.e. same at overall tendency but greatly different at local waveband. As shown in Fig.1, at seedling stage, the LAI and chlorophyll content was very small, and the reflectance of visible light (450~760nm) was higher than that at other growth stages due to mulching film. Thus, the canopy reflectance of visible light and near-infrared light was mainly influenced by the content of chlorophyll a/b, carotenoid and xanthophylls, and the optical characteristics of leaf tissue respectively. With growth progress, there was a continuous increase in tobacco biomass and chlorophyll content, and an increase in LAI. Canopy spectral reflectance decreased to a certain extent within visible light, increased rapidly within near-infrared area, peaked at vigorous growth stage, and gradually decreased at maturity stage with partial leaf harvest. 0.7 0.6
seedling stage
0.5 e c n e0.4 t c e l0.3 f e r 0.2
rosette stage vigorous stage maturity stage
0.1 0 400
500
600
700
800
900
100
wavelength(nm)
Fig. 1. Canopy spectral characteristic curves of tobacco at different growth stages
654
H. Qiao et al. 0.8 0.7 0.6 ec na 0.5 tc el 0.4 f er 0.3
N0 N3 N6 N9 N12
0.2 0.1 0 400
500
600
700 waveband
800
900
1000
Fig. 2. Canopy spectral characteristic curves of tobacco at different N levels(Pre-mature stage)
As shown by data at 4 growth stages, the spectral characteristics of tobacco canopy changed roughly at same tendency, but at significant difference for different nitrogen fertilizer levels (and very significantly even at early-maturity stage (Fig.2), i.e. P>0.05 at t test (760nm, 800nm, 850nm and 960nm respectively). 3.2 Correlation of Vegetation Index with Chlorophyll Content and LAI As shown by Table 2, LAI was very significantly related to RVI and NDVI (green); and significantly related to NDVI, SAVI and RDVI respectively. NDVI (green) was a spectral parameter closely related to leaf chlorophyll content and LAI. 3.3 Regression Equation of NDVI (Green) with Chlorophyll Content and LAI According to the NDVI (green), regression equation was established for NDVI (green) with LAI. As shown by regression equation with an independent variable of NDVI (green) (Table 3), a regression equation could be established for LAI. Determination coefficient (R2) was 0.568 significantly for the regression equation of LAI, and
Table 2. Correlation analyses between vegetable index and chlorophyll, LAI (n=90) Item
RVI
NDVI
NDVI(green)
SAVI
OSAVI
MSAVI
RDVI
0.426**
0.222*
0.484**
0-.227*
-0.062
-0.125
-0.242*
Associated Probability
0
0.018
0
0.08
0.575
0.147
0.023
Variation Coefficient
0.546
0.003
0.007
-0.004
0
-0.002
-0.005
Correlation Coefficient LAI
Table 3. Regression equations of chlorophyll and LAI by NDVI(green)
Dependent LAI
Regression eqution Y=0.689+0.035X1
* Significant(P 0.05)
Independent X1
The name of independent NDVI(green)
Regression coefficient Sig 0.024
R2 0.568*
Study on Relationship between Tobacco Canopy Spectra and LAI
655
regression coefficient was also significant. Thus, the established regression equation was effective, and NDVI (green) could be used to estimate LAI.
4 Conclusion and Discussion Nitrogen is necessary for the growth, development, quality and yield of tobacco. However, excessive nitrogen caused spindly growth, formed black tobacco leaf, and greatly lowered tobacco quality. LAI is important index of tobacco photosynthesis and metabolism, and closely related to final yield and quality. Thus, the fertilization state of field and the growth information of tobacco must be obtained rapidly to support in time the decision on production management. Hyperspectral remote technique is a determination technique to rapidly obtain the growth state of crop. This study preliminarily described the spectral characteristics of tobacco canopy under different fertilities at different fertilization levels at different growth stages, analyzed the correlation of vegetation index with LAI based on canopy spectrum, and established estimation model. Under different fertilization conditions, the spectral reflectance of tobacco canopy was significantly different, NDVI (green) was significantly related to LAI, and the regression coefficient, association probability and determination coefficient were all significant for established regression. Thus, the screened variable of hyperspectral characteristics was reliable, and relevant model could be used to estimate chlorophyll content and LAI. In the past, many studies were made on wheat, maize and rice to screen corresponding characteristic variable and statistical model [5, 11, 12] . The ratio of near-infrared light to green/red light could significantly improve the sensitivity to parameters of canopy structure. The information of near-infrared light could standardize the established vegetation index, eliminate the influence of canopy structure as far as possible (such as leaf structure, leaf direction and radiation angle), and effectively highlight the information of sensitive waveband (such as green/red light) [13]. In this study, the screened NDVI (green) was also based on vegetation index of near-infrared and green lights, and proven also effective for tobacco. In wheat study of Aparicio et.al. [14], the correlation of LAI with vegetation index was not affected by planting measures, species or planting area. But such results on tobacco should be verified through further test. In this study, SPAD-502 chlorophyll detector was used to rapidly harmlessly determine the chlorophyll content in tobacco. In soybean/maize study of John et.al. [15], the actual amount of chlorophyll extract was very significantly related to the detected data through SPAD chlorophyll detector (R2=0.94). Such results appeared in the study of Carlos et.al. [16]. In the study of Zeng Jianmin et.al. [17], tobacco chlorophyll was extracted through 95% water solution of acetone and anhydrous ethanol (2:1 volume); the chlorophyll content in leaf and SPAD value were at very significantly positive correlation; and the chlorophyll content in leaf was estimated through SPAD value to further track the nitrogen nutritional state in tobacco. In this study, canopy hyperspectral monitored the chlorophyll content and LAI tobacco, rapidly effectively monitored the growth state of tobacco, and provided technological support for tobacco planting and fertilizer management. Of course, the fitting
656
H. Qiao et al.
equation should be further verified through the tests at different ecosystems and different species, so as to popularize the model and provide theoretic support for large-scale monitoring.
Acknowledgements This work was financially supported by the National Key Technology Research and Development Program of China (No. (2006BAD08A01)) and Henan Tobacco Company Projects Grant (HNKJ200814).
References 1. Shen, G.R., Wang, R.C.: Review of the application of vegetation remote sensing. Journal of Zhejiang University (Agric. &Life Sci. ) 27(6), 682–690 (2001) (in Chinese) 2. Curran, P.J.: Remote sensing of foliar chemistry. Remote sensing of environment 30, 271–278 (1989) 3. Hinzman, L.D., Bauer, M.E., Daughtry, C.S.T.: Effects of nitrogen fertilization on growth and reflectance characteristics of winter wheat. Remote sensing of environment 19, 47–61 (1986) 4. Svetlana, M.K., Taras, A.K.: Changes in the first derivatives of leaf reflectance spectra of various plants induced by variations of chlorophyll content. Journal of Plant Physiology 12(3), 1648–1655 (2007) 5. Daughtry, C.S.T., Walthall, C.L., Kim, M.S., et al.: Estimating Corn Leaf Chlorophyll Concentration from Leaf and Canopy Reflectance. Remote Sensing of Environment 74(2), 229–239 (2000) 6. Elizabeth, J.B., Brigitte, L., Bernie, Z., et al.: Non-destructive estimation of potato leaf chlorophyll from canopy hyperspectral reflectance using the inverted PROSAIL model. International Journal of Applied Earth Observation and Geoinformation 9(4), 360–374 (2007) 7. Bouman, B.A.M.: Linking physical remote sensing model with crop growth simulation models, applied for sugar beet. International of Remote Sensing 13(2), 2565–2581 (1992) 8. Wu, J.D., Wang, D., Marvin, E.B.: Assessing broadband vegetation indices and QuickBird data in estimating leaf area index of corn and potato canopies. Field Crops Research 102(1), 33–42 (2007) 9. Hung, T.N., Byun, W.L.: Assessment of rice leaf growth and nitrogen status by hyperspectral canopy reflectance and partial least square regress. European Journal of Agronomy 24(4), 349–356 (2006) 10. Liu, G.S.: Tobacco Cultivation, pp. 38–39. China Agriculture Press, Beijing (2003) 11. Feng, W., Yao, X., Zhu, Y., et al.: Monitoring Leaf Nitrogen Concentration by Hyperspectral Remote Sensing in Wheat. Journal of Triticeae Crops 28(5), 851–860 (2008) (in Chinese) 12. Vaesen, K., Gilliams, S., Nackaerts, K.: Ground-measured spectral signatures as indicators of ground cover and leaf area index: the case of paddy rice. Field Crops Research 69(1), 13–25 (2001) 13. Xue, L.H., Cao, W.X., Luo, W.H., et al.: Relationship Between Spectral Vegetation Indices and LAI In Rice. Acta Phytoecologica Sinica 28(1), 47–52 (2004) (in Chinese)
Study on Relationship between Tobacco Canopy Spectra and LAI
657
14. Aparicio, N., Villegas, D., Araus, J.L., et al.: Relationship between growth traits and spectral veg2etation indices in Durum wheat. Agronomy Journal 42, 1547–1555 (2002) 15. John, M., John, C.O., Jennifer, L.M.: Calibration of the Minolta SPAD-502 leaf chlorophyll meter. Photosynthesis Research 46, 467–472 (1995) 16. Carlos, C., Lianne, M.D., Pierre, D., et al.: Inter-relationships of applied nitrogen, SPAD, and yield of leafy and non-leafy maize genotypes. Journal of Plant Nutrition 24(8), 1173–1194 (2001) 17. Zeng, J.M., Yao, H., Li, T.F., et al.: Chlorophyll Content Determination and its Relationship with SPAD Readings in Flue-cured Tobacco. Molecular Plant Breeding 7(1), 56–62 (2009) (in Chinese)
Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI in Inner Mongolia Grassland Hongbin Zhang1,2,3, Guixia Yang1,2,3, Qing Huang2,3, Gang Li1,2,3, Baorui Chen1,2,3, and Xiaoping Xin1,2,3,* 1
Hulunber Grassland Ecosystem Observation and Research Station, Beijing, 100081, China 2 Key Laboratory of Resource Remote Sensing and Digital Agriculture, Ministry of Agriculture, Beijing, 10008, China 3 Chinese Academy of Agricultural Sciences Institute of Agricultural Resources and Regional Planning, Beijing, 100081, China
[email protected] Abstract. Based on MODIS NDVI and NOAA NDVI datum, covering the primary grassland types of Inner Mongolian in growing seasons from 2000 to 2003, this paper analyzes annual variation rule of the relationship between MODIS NDVI and NOAA NDVI datum. We use the theory of statistics to discuss the spatial scaling methods between different resolutions images of remote sensing in large-scale spatial extent. At the same time, we build spatial scaling model by MODIS NDVI and NOAA NDVI datum of July and August in 2002, and apply it to the 2003’s NOAA NDVI datum, then take the survey datum in field to validate the precision of the model. The result indicates that this spatial scaling method is effective, and the model could be applied to other times. This method makes it scientific and effective to analyze and compare the result of monitor by NOAA NDVI and MODIS NDVI of different times in grassland. Keywords: Spatial scaling, grassland remote sensing, MODIS, NOAA.
1 Introduction At present, remote sensing technology has already become the mainly acquisition method and important study means for macroscopic ecology subject. Because satellite sensor imaging has instantaneity and periodicity features coupled with weather factors (such as cloud, rain, snow and so on) affecting sensor imaging, any sensor can’t supply the enough image datum to covered the whole grassland in any times (Li Xin et al., 2007). That’s say, shortage of time series image in the same region or spatial *
Corresponding author. Address: Chinese Academy of Agricultural Science Insititute of Agricultural Resource and Regional Planning, No.12 Zhongguancun South St., Haidian District, Beijing 100081, China. Tel:+86-10-82109622-138,
[email protected].
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 658–666, 2011. © IFIP International Federation for Information Processing 2011
Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI
659
image data in the same time for macroscopical ecology system study. Especially in long time serial and large spatial scale ecology system monitoring study, this problem is the most serious. So how to effectively integrate multiple-source remote sensing datum and different resolution (spatial resolution, temporal resolution) image datum for macroscopical ecology monitoring study has become difficulties in current. Especially in recent years, many experts and scholars at home and abroad have already developed search. Mayaux and Lambin adopted two-steps transformation method to make vegetation area scale transformation based on four spatial exponential relations of TM and AVHRR in 1995(Mayaux P et al., 1995). Wang Kaicun and so on analyzed shortwave albedo of MODIS and AVHRR data and compared the differences between them in details in 2004(Wang kaicun et al., 2004). Kevin Gallo and so on analyzed differences of NOAA NDVI and MODIS NDVI in details and built up NOAA NDVI and MODIS NDVI data relational model of types of farmland, grassland, evergreen broad-leaf forest, shrubbery, city in 2005(Kevin Gallo et al., 2005). Bao Pingyong and so on analyzed surface albedo production of same time’s ETM+ and MODIS data and compared differences of surface albedo of forest land, farmland, grassland, water, city, river shoal, exposed soil and exposed rock in 2007(Bao PIngyong et al., 2007). Zhang Wanchang and so on took on scale transformation study based on LAI index inverted from ETM+ data and attempted to bring forward a new more effective scale transformation method based on NDVI unmixed pixels in 2008(Zhang Wanchang et al., 2008). Wang Peijuan adopted ETM+ and MODIS data to investigate spatial scale transformation method of forest coverage net primary productivity in Changbai mountain nature reserve (Wang Peijuan et al., 2007). This paper took Inner Mongolia grassland as an example to study spatial scale transformation method among large scale grassland different resolutions remote sensing data. Because MODIS and NOAA data have lower spatial resolution features (Huang Jiajie et al., 2003; Potte, C et al., 2003; Steven, M. D et al., 2003 ), and survey region is too large, it’s difficult obtain high precision land utilization type map and different types of vegetation pure pixels and so on parameters. So we used the theory of statistics to build spatial scaling model of MODIS NDVI and NOAA NDVI datum, and applied the model to time scale application.
2 Survey Region Overview Inner Mongolia grassland is located in hinterland of Eurasian Continent. It is a vast land and has rich and colorful grass types occupying 22% of our country grassland whole area. at the same time it spans “Sanbei distraction” in our country. It’s important animal breeding production base and a natural green protective screen for north china, it takes the role of protection and improvement our country ecology environment. According to Inner Mongolia autonomous region 1:10000000 grassland type map, whole grassland area is 794239.00 km2. The survey region makes up of Lowland meadow, up-land meadow, temperate steppe, steppe desert, temperate meadow steppe, desert grassland, desert steppe and so on seven types of grassland and whole area is 781587.10 km2 accounting for 98.4% of Inner Mongolia grassland whole area. It is the main body of Inner Mongolia grassland autonomous region.
660
H. Zhang et al.
Fig. 1. The Grass type of Inner Mongolia
3 Study Method 3.1 Data Resources This paper used the first and second band of MOD09Q1 production provided by NASA. Because MOD09Q1 is 8 days reflectivity production composed by everyday L2G surface reflectivity and affected by weather easily, it is very hard to get the remote sensing data which was not shaded by cloud in a short time (Yan Jianwu et al., 2008; Zhang Lianyi et al., 2008). So this paper applied the maximum value compounding method to obtain monthly NDVI vegetation index data to analyze. The NOAA data we used is 8 km resolution NOAA/AVHRR monthly maximum value compounding NDVI production provided by Global Inventory Monitoring and Modeling Studies. 3.2 MODIS NDVI and NOAA NDVI Data Relation Stability Analysis Because this study is involved with spatial scale transformation model applied in time scale, it is necessary to analyze stability of correlation between MODIS NDVI and NOAA NDVI data in long time scale. It is the basic premise to ensure stability and effectiveness of time scale transformation model. This paper selected monthly maximum compounding NOAA and MODIS NDVI data from April to October, in 2000 to 2003, and use ArcGIS 9.0 to calculate correlation of different months NDVImax between two types of data. Compare fluctuation of correlation in same term and study whether there is stable correlation between them in research area. The results as shown as below:
Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI
661
Fig. 2. Analysis of the correlation between NOAA NDVI and MODIS NDVI, 2000-2003
From fig 2, we can see correlation of NOAA NDVI and MODIS NDVI data changed with regularity and the correlation was stable in long time scale between them in growing season from 2000 to 2003 except 2001. 3.3 Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI Because MODIS sensor is superior to AVHRR sensor in spatial and spectral resolution, MODIS NDVI can reflect grassland vegetation condition more truly. So this paper selected down-scale transformation method that from NOAA NDVI to MODIS NDVI and built scaling model. The detailed method is as below. Apply systematic sampling to select study samples in MODIS NDVI and NOAA NDVI remote sensing images. Sampling interval is 40 km. Sampling range of MODIS image was an circular region with 4km radius, take NDVI mean value as MODIS data sampled value. The Sampling range of NOAA image was a pixel (8km*8km), so we can obtain 533 MODIS and NOAA sample datum every month in research area.
0.9
0.7
0.8
0.6
0.7
I0.5 V D N0.4 S I0.3 D O M 0.2
I0.6 V D N0.5 S I0.4 D O M0.3
y = 1.0303x + 0.0464 2
R = 0.64
0.1
0.2
y = 0.8101x + 0.0781
0.1
R = 0.7301
2
0.0
0.0 0.0
0.1
0.2
0.3 0.4 NOAA NDVI
0.5
0.6
0.7
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
NOAA NDVI
Fig. 3. Comparison of NOAA NDVI and Fig. 4. Comparison of NOAA NDVI and MODIS NDVI in Apr MODIS NDVI in May
662
H. Zhang et al.
1.0
1.0
0.9
0.9
0.8
0.8
I0.7 V D0.6 N S0.5 I D0.4 O M 0.3
y = 0.9557x + 0.0684
0.7 I V 0.6 D N 0.5 S I D O M 0.4 0.3
0.2
R = 0.8449
y = 0.9679x + 0.0491
0.2
2
0.1
0.1
0.0
0.0
2
R = 0.8726
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 NOAA NDVI
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 NOAA NDVI
Fig. 5. Comparison of NOAA NDVI and Fig. 6. Comparison of NOAA NDVI and MODIS NDVI in Jun MODIS NDVI in Jul 0.9 0.8
1.0 0.9 0.8 I0.7 V D N0.6 S0.5 I D0.4 O M0.3 0.2 0.1 0.0
0.7 0.6 VID N 0.5 SI ODM 0.4 0.3 0.2
y = 0.9906x + 0.0525
y = 0.9905x + 0.044 R2 = 0.8512 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 NOAA NDVI
2
R = 0.8776
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 NOAA NDVI
Fig. 7. Comparison of NOAA NDVI and Fig. 8. Comparison of MODIS NDVI in Aug MODIS NDVI in Sep
NOAA NDVI and
0.8 0.7 0.6 I V0.5 D N S0.4 I D OM0.3 0.2
y = 1.1524x + 0.0191 R2 = 0.7438
0.1 0.0 0.0
0.1
0.2
0.3
0.4 0.5 NOAA NDVI
0.6
0.7
0.8
Fig. 9. The Comparison of NOAA NDVI and MODIS NDVI in Oct
From fig 3 to 9, MODIS NDVI and NOAA NDVI data of grassland vegetation in April to October have well linear correlation which increase start at April, reached a peak in August and decrease start at September. As correlation was concerned, April was at lowest, May and October were almost the same, but October was slightly higher than May; June and September were almost the same, but September was
Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI
663
slightly higher than June; July and Aug were almost the same, but Aug was slightly higher than July; May and October were significantly higher than April; June to September were significantly higher than April, May and October; July and August were significantly higher than June and September. The change trend of linear correlation between MODIS NDVI and NOAA NDVI data was truly in response to grassland vegetation condition in research area (Xu Bin et al., 2007). The grassland began to growing in April and at this time, the grassland vegetation coverage and biomass were the lowest in growth season. So NDVI index of grassland was very small, when the influence of soil background noise etc on AVHRR and MODIS sensors monitoring relatively the largest, more mistakes resulted. Especially NOAA NDVI data, because of uncertainty of calibration parameter, rang of mid-value and low value excelled 20%. All of the facts disturbed MODIS NDVI and NOAA NDVI data to obtain the parameters reflecting grassland vegetation condition to a great degree. So NDVI data has too much noise in April and correlation is the worst. In May grassland vegetation coverage and biomass were better than April. The influence of factors of soil background noise etc on AVHRR and MODIS sensors monitoring began to weaken. So correlation of MODIS NDVI and NOAA NDVI rose notably compared with April. In June, grassland vegetation coverage and biomass had a further marked increase and correlation continued to strengthen. July and august were the two months in which grassland vegetation condition were the best in the whole growth season. So correlation of July and August were higher than the other months and grassland vegetation condition in July was better than that in August. But because saturation effect of AVHRR and MODIS sensor bothered to obtain NDVI data reflecting truly grassland vegetation condition, correlation of August was a little higher than that in July. In September, grassland vegetation coverage and biomass began to decrease, but because there was a lot of hay in research area which decreased the influence of soil background noise on AVHRR and MODIS sensor monitoring, association relationship of September was just a little higher than June. In like manner, correlation of October was just a little than May.
4 Spatial Scale Transformation Method Application and Verification The NDVI values of July and August represented the maximum value all the year round in research area. So NDVI value of July and August were applied the most widely in biomass yield estimation. Because of shortage of ground surface survey data except July and August, this paper mainly validated the precision of spatial scale transformation model between MODIS NDVI and NOAA NDVI data of July and August in 2002, according to the spatial scale transformation discussed in 3.3, linear spatial transformation model was built up and extrapolated to NOAA NDVI data of July and August in 2003. Taken use of MODIS NDVI and NOAA NDVI data and survey data to validate and verify spatial scale transformation precision when extrapolated to time scale and practicability. Specific process as follows: Adopt systematic sampling method, draw 533 samples in 8km resolution NOAA NDVI and 250 mm resolution MODIS NDVI of July and August in 2002. According to linear correlation, built up spatial scale transformation model of NOAA NDVI and
664
H. Zhang et al.
MODIS NDVI in July and August (table 1). Take advantage of this model and Erdas 9.0 to resample pixels (to 250 m) and linear process to obtain transformed 250m NOAA NDVI data of July and August in 2003. Use survey datum of July and August in 2002 and MODIS NDVI data of July and August in 2002 to build up regression model to get the regression formula: Y=11.2620+179.6915X (Y: dry biomass; X: MODIS NDVI). Input above transformed 250m resolution NOAA NDVI data to model and generate grassland vegetation biomass distribution map of July and August in 2003 separately. Use survey datum of July and August in 2002 and untransformed 8km resolution NOAA NDVI data to build up regression model to get the regression formula: Y=25.9050+129.6058X(Y: dry biomass;X:NOAA NDVI vegetation index). Input above uncorrected 8km resolution NOAA NDVI data in 2003 to model and get ground biomass distribution map in research area of July and August in 2003. Table 1. Spatial Scaling Model of MODIS NDVI and NOAA NDVI in 2002
Spatial Scaling model
R2
July
Y=0.028442+0.975085X
0.93
Sum of Squared Residuals 2.22
August
Y=0.027654+0.983479X
0.93
2.29
:Y:MODIS NDVI X:NOAA NDVI
Note
At last, take advantage of survey datum of July and August in 2003 (29 ground surface samplings in July and 52 ground surface samplings in August) to verify biomass distribution map and analyze precision difference. The results as follows: Table 2. Result of Verify Spatial Scaling Model of NOAA NDVI and Biomass, Jun and Jul 2003 Mean relative bias Δ
NOAA NDVI Untransformed July data in 2003 Transformed July data in 2003 Untransformed August data in 2003 Transformed August data in 2003 Untransformed July and August data in 2003 Transformed July and August data in 2003
:mean relative bias: Δ
Note
=
∑
d
/ x
i = 1
V
=
0.36 0.35 0.28 0.27
bias V 38.87 40.44 47.34 42.08
0.31
44.49
0.29
41.75
n
(
∑
i = 1
i
) / n
;mean absolute bias:
2
n
(
Mean absolute
d
) / n
,x is measured value,n is sampling quantity
d is the difference of predicted value and measured value
Results of verify in table 2 indicated that NOAA NDVI in 2003 transformed from 2002 by spatial scale transformation model had very high monitoring precision to
Study on Spatial Scale Transformation Method of MODIS NDVI and NOAA NDVI
665
grassland biomass and had improved compared with untransformed data. So the spatial scale transformation model can extrapolated in time scale.
5 Conclusions and Discussion 1. On the whole, MODIS NDVI was more than NOAA NDVI in value, and they both had high relativity in growth season. The relativity mainly affected by vegetation growth climate condition. That is because when climate condition was better, high coverage of grassland can reduce noise influence of soil background on AVHRR and MODIS sensors imaging. On the contrary, when grassland vegetation condition was bad, noise influence of soil background on AVHRR and MODIS sensors imaging would be increased, resulted correlation of MODIS NDVI and NOAA NDVI decreased. So in research area, correlation of July and August were the highest and April was the lowest. 2. Linear spatial transformation model built up based on MODIS NDVI and NOAA NDVI was stable in time dimension in some degree. After time scale transformation model was extrapolated application, the transformed NOAA NDVI data kept higher monitoring precision of grassland biomass to indicate the model could be extrapolated in time scale. Because vegetation condition in July and August were the best, association degree between MODIS NDVI and NOAA NDVI data were the highest. So stability of spatial scale transformation model built up based on MODIS NDVI and NOAA NDVI in July and August was the best when extrapolated in time scale. 3. This paper based on statistics method, making use of association relationship of 250m resolution MODIS NDVI and 8 km resolution NOAA NDVI data, spatial transformation model between MODIS NDVI and NOAA NDVI data was built up. But because this method self did not take 8 km resolution NOAA NDVI data big pixel inner heterogeneity in to account. So relative precision of 250mm resolution small pixel of NOAA NDVI data in 8 km large pixel region was not prompted and this was the shortcoming of the method. In so large spatial scale survey region, conflicted by so many basic data serious limits (such as high precision land utilization map shortage, big mistake in spatial adjustment of low resolution data, different type vegetation pure pixel shortage) It was difficult to adopt other scale transformation method taken spatial heterogeneity into account. Under the premise of ensuring total scale transformation precision, the method adopted by this paper was a scale transformation feasible scientific method in large space range different resolutions remote sensing data.
Acknowledgements This work is supported by Special Fund Project for Basic Science Research Business Fee, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences and The National Science & Technology Program (Grant No. 2006BAC08B0404 、 2007BAC03A10) and Project 863 of China: (Grant no.2007AA10Z230) and National Natural Science Foundation of China (Grant No: 30770327) and Commonweal Industry Scientific Research Special Funds Project (GYHY200906029-2).
666
H. Zhang et al.
References 1. Li, X., Huang, C., Che, T., et al.: Progress and Foresight of Chinese land Data Assimilation System Study. Nature Science Progress 17(2), 163–173 (2007) 2. Mayaux, P., Lambin, E.F.: Estimation of tropical forest area from coarse spatial resolution data: A two-step correction function for proportional errors due to spatial aggregation. Remote Sensing of Environment 53(1), 1–15 (1995) 3. Wang, K., Liu, J., Zhou, X., et al.: Retrieval of the surface albedo under clear sky over China and its characteristics analysis by using MODIS satellite date. Chinese Journal of Atmosphere Sciences 28(6), 941–949 (2004) 4. Gallo, K., Ji, L., Reed, B., Eidenshink, J., Dwyer, J.: Multi-platform comparisons of MODIS and AVHRR normalized difference vegetation index data. Remote Sensing of Environment (99), 221–231 (2005) 5. Bao, P., Zhang, Y., Gong, L., et al.: Study on consistency of land surface albedo obtained from ETM+ and MODIS. Journal of Hehai University (Natural Sciences) 35(1), 67–71 (2007) 6. Zhang, W., Zhong, S., Hu, S.: Spatial scale transferring study on leaf area index derived from remotely sensed data in the Heihe river basin, China. Acta Ecologica Sinica 28(6), 2495–2503 (2008) 7. Wang, P., Xie, D., Zhang, J., et al.: Spatial scaling of net primary productivity based on process model in Changbai Mountain natural reserve. Acta Ecologica Sinica 28(8), 3215– 3223 (2007) 8. Huang, J., Wan, Y., Liu, L.: The character and application of MODIS. Geospatial Information 1(4), 20–28 (2003) 9. Potte, C., Tan, P.N., Steinbach, M., Klooste, S., Kumar, V., et al.: Major disturbance events in terrestrial ecosystems detected using global satellite data sets. Global Change Biology (9), 1005–1021 (2003) 10. Steven, M.D., Malthus, T.J., Baret, F., Xu, H., Chopping, M.J.: Intercalibration of vegetation indices from different sensor systems. Remote Sensing of Environment (88), 412–422 (2003) 11. Yan, J., Li, C., Yuan, L., Chen, Q.: Application summary of EOS-MODIS data in the monitoring of grassland resources. Pratacultural Science 25(4), 1–9 (2008) 12. Lianyi, Z., Gang, W., Luru, B.: Temporal changes of MODIS-NDVI vegetation index and forage biomass in Xilinguole grassland- taking the change from April to September in 2005 as a sample. Pratacultural Science 25(3), 6–11 (2008) 13. Xu, B., Tao, W., Yang, X., et al.: Monitoring by remote sensing of vegetation growth in the project of grassland withdrawn from grazing in countries of China. Acta Prataculturae Sinica 16(5), 13–21 (2007)
Study on Storage Characteristic of Navel Orange Based on ANN Junfang Xia and Runwen Hu College of Engineering and Technology Huazhong Agricultural University, Wuhan 430070 China
Abstract. In order to predict storage life of navel orange, The model for the variable regularity of total soluble sugar, total acidity, vitamin C, soluble solids, the sugar-acidity ratio in navel orange according to storage time was established based on BP artificial neural network . The results show that the multi-factor BP artificial neural network model has better predicted effect than single-factor one. When the number of the hidden layer neuron is 8, the multi-factor BP artificial neural network model of total soluble sugar, total acidity, vitamin C, soluble solids, the sugar-acidity ratio according to storage time was the most accurate, the correlation coefficient R between prediction and true value of storage time reached 0.98, the prediction and true value of the model was 0.99. As a result, the multi -factor BP artificial neural network model could be used to predict the navel orange storage life. Keywords: Navel orange; storage life; BP artificial neural network.
1 Introduction The quality of navel orange such as total soluble sugar, total acidity,VC,soluble solids, the sugar-acidity ratio will change along with storage. Grasping this regulation, finding out notable quality index related to storage and establishing prediction model about quality can provide the theoretical basis for predicting storage life of navel orange, control internal quality effectively, prevent deterioration of fruit, preserve the value of navel orange and make appropriate market decision in time. Variation of orange quality is nonlinear according to the time of storage, and the traditional linear regression model is difficult to build such a nonlinear system. Artificial neural network method is to simulate the thinking way of human being based on the working principle of human brain cells, so it can fully approach any complicated nonlinear relationship, and has powerful potential to solve the system control of highly nonlinear and serious uncertainty [1-6]. In order to establish artificial neural network model, Rumelhert et al presented BP Algorithm Used in Multilayer Neural Network[7], BP artificial neural networks are mostly applied in artificial neural network, The algorithm got very extensive application in nonlinear regression, Most of studies are about the BP artificial neural network [8] . For example, Liu Jianxue et al[9] used BP neural network method to establish protein D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 667–673, 2011. © IFIP International Federation for Information Processing 2011
668
J. Xia and R. Hu
prediction model of different types and sizes of rice samples, investigated the prediction ability of the model, and reported that there were good correlation (with the correlation coefficients above 0.9) between the predictive value and the value of chemical analysis using traditional methods. Lin Ming,Zhao Chen,Liu Xuesong,Bai Yingkui,Tang Yanfeng,Yang Nanlin et al determined the contents of corn, Cordyceps amino acids, Amino Acids in Cordyceps Sinensis,vanilla activity, VC Yinqiao Tablets and Anthraquinone in Rhubarb by the method of Artificial Neural Network- near-infrared spectroscopy. The results showed that the method was an effective and practical method of non-linear correction method, and the standard deviation of their forecasts are better than the processing results by principal component regression and partial least squares regression linear models[10-15]. In this study, the variety model of quality of navel orange along with storage time is based on artificial neural network, which helps to forecast the storage time and storage life by the change of quality of navel orange.
2 Materials and Methods 2.1 Materials The test samples are 200 mature navel oranges picked randomly from many trees in Dingnan in Jiangxi province in December 7, 2007, encapsulated by plastic film Bag, refrigerated in the artificial climate box with the temperature of 5°C and humidity of 70%. At interval of 10 days, randomly choosed 10 navel oranges were determined total soluble sugar, total acidity, vitamin C, soluble solids and the sugar-acidity ratio, the calculated average value and the storage time(day) were uesd as input vector P and output vector T of neural network respectivly. 160 of the 200 samples were used as training set and 40 as validation set. 2.2 Chemical Analysis The measurement of the soluble solids content in navel oranges was carried out by WYT-1 handhold saccharimeter; the contents of total soluble sugar were determined by phenol sulfuric acid method, the total acid contents were detected by aid-base titration method, and the method of 2,6-dichlorophenol indophenol was used to measure the contents of vitaminC. 2.3 Determination of Storage Life The observation of the stored orange showed that Mildew spots appeared on the skin of a small number of orange samples at 170 days. At 180 days, black spots appeared on the skin of about 20% of the navel oranges, this phenomenon indicated that the navel oranges had been deteriorated and could not be used as food. Therefore, the critical value of storage life of navel orange should be 170 days. This article will find index of internal quality related with the storage time significantly and estabilish the correlation
Study on Storage Characteristic of Navel Orange Based on ANN
669
model of characteristic of navel oranges to storage time so as to predict the storage time of navel orange. the continued storage time could be got by critical value of storage time subtracts the storaged time, that is storage time. 2.4 Structure of BP Artificial Neural Network The multi-player feed-forward neural network of directional transfer algorithm was applied to BP artificial neural network which contained input layer, hidden layer and output layer. In this study, both of the structural input layer and output layer of BP artificial neural network model for quality characteristics of navel orange to storage time (single factor) are unit-neuron. Input layer neuron is the internal quality index, the output layer neuron is the storage time. As shown in Fig.1,The input layer of BP artificial neural network model of multi-factor quality of navel orange according to storage time is 5 neurons, output layer is single neuron,as shown in Fig.2.
Fig. 1. The BP artificial neural network of single Fig. 2. The BP artificial neural network of factor multi factor
2.5 Network Training Parameter In this study, Levenberg-Marquardt algorithm was used. S-transfer function was adopted to hidden layer neurons, the transfer function was tansig function; The output layer was cell neuron, linear transfer function was used in neuronal function and trainlm function was used in training function to generate the neural network of initialization. Before the training, the sample must be ascertained and should includ input P and corresponding output T. During the training procedure, weight and threshold should be adjusted constantly to make the average variance mas of neural networks output and expection output vector to be the minimum. The definition of network training
670
J. Xia and R. Hu
parameters: the max training times epochs was set as 200; the training iterative process show as 100; this training required the accuracy as 1e-10; learningrate lr was demanded 0.01;the min gradient as 1e-6; the max failure times must be under 5[7].
3 Data Processing and Analysis 3.1 Establishment of Single-Factor Artificial Neural Network Model Internal quality indexes such as total soluble sugar content, total acidity, vitamin C, soluble solids and the sugar-acidity ratio of navel orange was taken as the input vector P and the storage time as expected output vector T of neural network. According to the programming and neural network training of definite network training parameters in the software MATLAB6.5, the quality to storage time of single-factor BP artificial neural network model for navel orange was established. The number of hidden layer neuron was initiativly set as 5 with the method of trial and error. Begainning with smaller neuron,we trained and inspected the capability of internet. As the number of neuron increased gradually, the capability of internet was trained and inspected again untill the the related coefficient Rc of output and expect value of artificial neural network model reached the ideal value. Right now, the number of the neuron of hidden layer is perfect. The characteristics of total soluble sugar content, total acidity, vitamin C, soluble solids and the sugar-acidity ratio of the 40 untraining samples were put into the neural network model established to predict the storage time. the conservasion of the established model will be analysed through comparing with true value. The best hidden layer in the BP artificial neural network model for total soluble sugar content, total acidity, vitamin C, soluble solids, the sugar-acidity ratio of navel orange according to storage time using was in table 1 as well as the effect on prediction and the proven results. In table 1,N is the number of best hidden layer, Rc is the related coefficient of predicted value of calibration model and measured value, Rp is the related coefficient of predicted value of validation model and measured value, Wp is the relative error of predicted value of the validation model and measured value. Table 1. Predictive results of the single factor BP artificial neural network model
Total soluble sugar Total acidity Vitamin C Soluble solids Sugar acid ratio
N 60 50 50 30 60
Rc 0.864 0.984 0.82 0.933 0.89
Rp 0.88 0.9814 0.8648 0.9343 0.9
Wp(%) 7.34 0.84 7.99 7.32 9.01
The model’s training research indicated that the number of neura in hidden layer of artificial neural network model had much influence to model’s accuracy and was accordng to different quality index of hidden layer of artificial neural network model.
Study on Storage Characteristic of Navel Orange Based on ANN
671
The model’s training time is related to the number of hidden layer neuron, the more of the number of hidden layer neuron,the longer of the training time; Among all model trainings,when the training time epochs was about 20,the mean square error of output layer reached the target error that training parameters demanded,and kept stable, not changed with epochs’s increasing. In the single-factor BP artificial neural network mode of navel orange’s internal qualities, total soluble sugar, total acidity, vitamin C, soluble solids and the sugar-acidity ratio according to storage time in BP artificial neural network mode were respectivly 60, 50, 50, 30, 60. The prediction accuracy and stability of the navel orange’s internal qualities to storage time in BP artificial neural network model with signal factor from high to low in sequence were total acidity BP artificial neural network model→soluble solids BP artificial neural network model→sugar-acid ratio BP artificial neural network model→total soluble sugar BP artificial neural network model→BP artificial neural network model of vitamin C. The result showed that the most obvious criterion of quality related to storage time in signal factor is total acidity.When the quantity of hidden layer neuron is 50,the total acidity to storage time of BP artificial neural network model accuracy is perfect,the related coefficient Rc can reach 0.984,the related coefficient Rp of validation model’s predicted value and measured value is 0.9814,the average value of relative error is 0.84%,the predictive effect is much better, so the neural network model established is stable, and can satisfy the prediction requirment of navel orange’s storage life. 3.2 The Establishment of the Multi-factor BP Artificial Neural Network The five indexes such as total soluble sugar content, total acidity, vitamin C, soluble solids, sugar-acid ratio of navel orange are taken as input vector P of neuron network, the number of input layer neuron is 5 and the storage time is used as expected output vector T. According to the definite network training parameters, programming and network training in the software MATLAB6.5, we established the multi-factor BP artificial neural network model. The method of trial and error was adopted to ascertain the optimum value of the hidden layer neuron’s quantity. After input the multi-factor indiex of the 40 navel oranges’ internal qualities untraining into the established neural network model and predict its storage time, the stability of the established model was analysed according to the comparison with measured value. The reaserch of model training indicates that the multi-factor BP artificial neural network dynamic model by the 5 internal qualities of total soluble sugar content, total acidity, vitamin C, soluble solids, sugar-acid ratio of navel orange according to the storage time is established ,when the number of hidden layer neuron is 8, the model accuracy is much high, the correlation coefficient Rc reach 0.98,the training time is shorter, the correlation coefficient of validation model’s predicted value and measured value Rp is 0.99,the average value of relative error is 6.701%, predictive effect is much better,the accuracy and stability of the established neural network model is better than single factor model which indicates that multi-factor changes can better reflect the variation of navel orange’s internal qualities to the storage time.using the multi -factor
672
J. Xia and R. Hu
BP artificial neural network model to forecast the storage time and storage life of navel orange is appropriate.
4 Conclusion The variety model of total soluble sugar, total acidity, vitamin C, soluble solids, the ratio of total soluble sugar to total acidity in navel orange according to storage time was established based on BP artificial neural network.. In the training of BP artificial neural network model, through changing the quantity of hidden layer neuron, the optimized number of the hidden layer neurons of the internal qualities of navel orange of BP artificial neural network with single-factor and five multi-factor was N, and verify the established model.The research results as follows. (1) The best forecast accuracy and stability of navel quality to storage time in single-factor BP artificial neural network model is total acidity of BP artificial neural network model , Therefore,in the single factor ,the index of quality which highlightly related to storage time is the total acidity of navel orange.And when the quantity of hidden layer neuron is 50,the accuracy of total acidity and storage time’s BP artificial neural network model is much higher, the correlation coefficient Rc reach 0.984,the correlation coefficient Rp of validation model’s predicted value and measured value is 0.9814,the average value of relative error is 0.84%. (2) When the number of hidden layer neuron is 8, the BP artificial neural network dynamic model which based on the five quality characteristies of total soluble sugar, total acidity, vitamin C, soluble solids and sugar acid ratio has the best prediction. The correlation coefficient Rc can reach 0.98, correlation coefficient Rp of validation model’s predicted value and measured value is 0.99, the average value of relative error is 6.701%. (3) The association between multi-factor changes and storage time is most notable,the predictive validity and stability of multi-factor BP artificial neural network model along with storage time is better than single-factor model. it indicates that multi-factor changing effects can best reflect navel orange’s qualities’ variety. In the practical application, multi-factor BP artificial neural network model can be used to predict the storage time and storage life of navel orange.
References [1] Guolin, W., Congming, H.: Application of neural network ensemble based on contractive algorithm in NIR analysis. Chinese Journal of Spectroscopy Laboratory 22(3), 473–476 (2005) [2] Yanbin, W.: Application of artificial neural network in NIR analysis and Dark-colored oil analysis. Research Institute of Petroleum Processing A Dissertation for PH.D (August 2000) (in Chinese) [3] Yong, Y., Ming, Q., Yunhong, L., et al.: Identification of alcohol quality via neural network based on genetic algorithms. Transactions of The Chinese Society of Agricultural Machinery 34(6), 104–106 (2003) (in Chinese)
Study on Storage Characteristic of Navel Orange Based on ANN
673
[4] Guangjun, Z.: Applications of artificial neural network to opto-electric measurements. Journal of Beijing University of Aeronautics and Astronautics 27(5), 554–568 (2001) (in Chinese) [5] Changhong, D., Matla, B.: Applications of neural network. National Defence Industry Publishing, Beijing (2005) (in Chinese) [6] Fuqiang, L., Wencui, Z., Ge, L., et al.: Nondestructive quality control of rutin pharmaceuticals by near infrared reflectance spectroscopy and artificial neural network. Chemical Analysis and Meterage 12(3), 11–13 (2003) (in Chinese) [7] Fei Sike Technology R & D Center. MATLAB 6.5 Neural Network Analysis and Design of Auxiliary. Electronics Publishing, Beijing (2005) (in Chinese) [8] Xiaoming, Q., Luda, Z., Xiaolin, D., et al.: Quantitative analysis using NIR by building PLS-BP model. Spectroscopy and Spectral Analysis 23(5), 870–872 (2003) (in Chinese) [9] Jianxue, L., Shouyi, W., Ruming, F.: Determination of protein content of rice by near infrared spectroscopy based on neural networks. Journal of Jiangsu University 25(3), 196–198 (2004) (in Chinese) [10] Ming, L., Jin, L.: Determination on components of corns based on neural networks and near infrared spectrum. Infrared Technology 26(3), 78–81 (2004) (in Chinese) [11] Chen, Z., Haibin, Q., Yiyu, C.: A new approach to the fast measurement of content of amino acids in cordyceps sinensis by ANN-NIR. Spectroscopy and Spectral Analysis 24(1), 50–53 (2004) (in Chinese) [12] Xuesong, L., Haibin, Q., Yiyu, C.: Determination of Active Components in a Natural Herb with Near Infrared Spectroscopy Based on Artificial Neural Networks. Chem. Res. Chinese. U 21(1), 36–43 (2005) [13] Bai, Y., Shen, X., Ding, D.: Two-component nondestructive analysis of VC Yinqiao tablets with NIR and bp neural network. Laser & Infrared 34(5), 354–356 (2004) (in Chinese) [14] Yanfeng, T., Zhuoyong, Z., Guoqiang, F.: Identification of official rhubarb samples based on NIR spectra and neural networks. Spectroscopy and Spectral Analysis 24(11), 1348–1351 (2004) (in Chinese) [15] Nanlin, Y., Yiyu, C., Haibin, Q.: Quantitative determination of mannitol in cordyceps sinensis using near infrared spectroscopy and artificial neural networks. Chinese Journal of Analytical Chemistry 31(6), 664–668 (2003) (in Chinese)
Study on the Differences of Village-Level Spatial Variability of Agricultural Soil Available K in the Typical Black Soil Regions of Northeast China Weiwei Cui and Jiping Liu* Tourism and Geographical Science College of Jilin Normal University, 136000, Siping, China
[email protected],
[email protected] Abstract. The spatial variability of soil nutrient is very important to the application of fertilizer, the sampling density of precision agriculture, and the sub-area of residence management of precision agriculture etc. With the examples of the agricultural soils of No.13 Village of Gongpeng Town and Xiguan Village of Enyu Town in Yushu City which are the typical black soil regions of Northeast China, applying semi-variance model and spatial autocorrelation model combined with GIS technology, a research on the differences of village-level spatial variability of agricultural soil available K is carried out in the thesis, which shows that the differences of averages and coefficients of variation in the same village are small, but large between different villages. The values of nugget and sill are close to the maximum ranges in No. 13 Village, but they are relatively different between different villages. The spatial autocorrelation of No. 3 and No. 7 lands in No. 13 Village are stronger than that of Xiguan Village. Keywords: Available K; Spatial Variability; Village-level Differences; Black Soil Regions of Northeast China; Precision Agriculture.
Researches on spatial availability of soil nutrient and management technology of precision agriculture have been very hot and have progressed rapidly in recent years [1]. With the putting forward and development of precision agriculture, the research on spatial variability of soil nutrient has already become one of the hotspot concerning modern soil sciences [2, 3]. The spatial variability of soil nutrient is very important to the application of fertilizer, the sampling density of precision agriculture, and the subarea of residence management of precision agriculture, which is the most primary and foremost issue for the research of precision agriculture and has a great scientific meaning and research value. There are fairly a number of researches on spatial variability of soil nutrient related to small-scale lands of plot-level and large-scale lands (e.g. countylevel) [2, 4 -15], but rather few related to the lands of village-level [12], and even fewer *
Corresponding author. Foundation Item: Jilin Province Science and Technology Support Program (20080207); Jilin Normal University Graduate Innovative Research Projects (S09010125).
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 674–681, 2011. © IFIP International Federation for Information Processing 2011
Village-Level Spatial Variability of Agricultural Soil Available K
675
related to the lands under the same natural conditions both at home and abroad. Taking the spatial variability of soil available K as an example, by applying statistical analysis technique and geographic information system (GIS) technology, the differences of village-level spatial variability of agricultural soil nutrients are researched which supplies a scientific basis for the management and rational fertilizing of precision agriculture, meanwhile, the research on the spatial-temporal variability of soil available K lays a foundation for the scientific management of soil available K and rational fertilizing [16]. Through the spatial statistical analysis which takes sampling point as basic information resource, a theoretical optimized soil sampling density is put forward, and then, combined with an economic and rational consideration, a feasible soil sampling density will be made, which is one of the effective methods for designing a soil sampling plan [17]. Geo-statistics is a relatively good method to study the spatial variability characteristics of soil property, and has been widely applied in recent years [18]. An analysis combined by applying GIS can further clarify the situation of soil available K of the land, reflect the law of spatial variability of soil available K, and offer necessary means to explain the spatial distribution characteristics of soil available K.
1 Research Regions and Methods 1.1 An Introduction to the Region Yushu City is a typical black soil region of Northeast China, which, thus, is selected as the research region in the thesis. Yushu City lies in the North Central Jilin Province, which is the center of the triangle district formed by three big cities: Changchun, Jilin and Harbin, with 30 towns, 4 sub-district offices, 388 villages, a population of 122 ten thousand, a total area of 4722 km2, and an agricultural acreage of 290,700 hm2 covering 68% of the area of the whole city, and is a typical black soil region. Three agricultural plots in Gongpeng Town and Enyu Town of Yushu City are taken as research objects (Figure 1). Gongpeng and Enyu towns are two neighboring towns which have similar natural conditions. The research regions belong to the temperate zone which is sub-humid and mild, with clear four seasons, sufficient sunlight (percentage of sunlight amounts to 60%) and an average annual sunshine of 2800h, an effective accumulated temperature of 2800 , the annual lead wind direction of southwest with a maximum speed of 3m/s, and an average annual rainfall of 620mm, which offer preferable weather conditions to the growth of crops and cash crops.
℃
1.2 Research Methods Sampling is carried out in No. 3 and No. 7 lands of No. 13 Village in Gongpeng Town and in No. 9 Land of Enyu Town in the thesis (Figure 1), and the sampling interval is 40m×40m, according to Quincunx Sampling Method, 72 samples are taken from each land (12 rows time 6 columns), and there are 216 samples taken in total. And then, extractions with the help of ammonium acetate—flame photometry will be applied to test the content of available K in these samples. A routine statistical analysis will be made on the data of soil available K of each plot through SPSS15.0, the
676
W. Cui and J. Liu
Fig. 1. Location Diagram of Research Regions
semi-variance model of each plot will be worked out by applying GS+ software, and then a contrastive analysis will be carried out, with the help of GeoDa software, the overall Moran’s I index of soil available K of all plots will be figured out and then be checked.
2 Results and Analyses 2.1 Analysis of the Characteristics of Descriptive Statistics A descriptive statistical analysis is carried on towards the soil available K of each plot by applying SPSS15.0 in the thesis (Table 1). The average value of soil available K changes a lot between different plots, but the average value of No. 3 Land is not much different from that of No. 7 Land in No. 13 Village, but there is a great difference between these two averages and that of No. 9 Land of Xiguan Village. The variation coefficient reflects the relative degree of variation of the variable quantity, which is small between No. 3 Land and No. 7 Land in No. 13 Village, but there is a great difference between them and that of No. 9 Land of Xiguan Village. Soil belongs to a natural continuum, and spatial variability is a kind of natural attribute of it [19]. The variation degree of soil depends on its forming process and its balance between time and space. Due to the influences of some factors like human activities, soil available K has the nature of spatial variability. For the
Village-Level Spatial Variability of Agricultural Soil Available K
677
reason that peasants still fertilize according to their experience, and the peasants in the same area, due to the influence of primary farming experience, fertilize the same quantity and type of fertilizer in the same land, which leads a relatively great difference between different villages. Table 1. Descriptive Statistical Analysis of Each Plot Item
No. 9
No. 3
No. 7
72
72
72
Max
295.3
167
171
Min
215.8
136
137
Average
253.943
152.36
157.22
Std.
19.3104
6.488
6.059
Mea.
0.29
-0.378
-1.147
Kurtosis
-0.105
-0.45
1.899
Cov.
0.076
0.043
0.039
No.of Samples
(Std.: Standard Difference;Mea.:Measure of Skewness;Cov.: Coefficient of Variation) 2.2 Selection and Analysis on Semi-variance Model The semi-variance model of each plot is worked out by applying GS+ software (Table 2, Figure 2). By adopting the semi-variance structure of soil available K calculated through GS+ software, and analyzing on the residual errors of all models, we finally choose the optimal one to further launch an analysis on spatial variability. With respect to the determination coefficient (Table 2), those of the two mentioned lands in No. 13 Village are much larger than that of No. 9 Land of Xiguan Village. The ratio between nugget and sill indicates the degree of spatial variability, and if the ratio is high, the degree of spatial variability caused by the stochastic part is relatively great; otherwise, the degree of spatial variability caused by spatial autocorrelation part is relatively great [1]. The relatively high ratio between nugget and sill in Xiguan Village illustrates that the degree of spatial variability caused by the stochastic part is relatively great, which shows that human factors play a leading role in the influences on the spatial variability of soil available K; while, the relatively low ratios between nugget and sill of No. 3 Land and No. 7 Land in No. 13 Village illustrate that the degree of spatial variability caused by spatial autocorrelation part is relatively great, and natural factors have much greater effects on No. 13 Village. The maximum correlation, namely distance refers to the variable function; from which we can see that the models of No. 3 Land and No. 7 Land in No. 13 Village are relatively close to each other, while the range of Xiguan Village is relatively different from those of the two lands in No. 13 Village.
678
W. Cui and J. Liu Table 2. Semi-variance Model and Its Fitted Parameters
Plot
Model
Nug.
Sill
No. 3
Glo.
13.7
46.98
No. 7
Exp.
9.42
No. 9
Lin.
299.8
Nug.
Max
C De.
RE.
0.29
199.3
0.984
10.25
34.71
0.27
139.8
0.837
20.5
430.22
0.70
396.13
0.558
8987
(Glo.: Linearity;Exp.: Exponential; Lin.:Globular; Nug.:Nugget ; Nug.S:Nugget/Sill; C De.:Coefficient of Determination ;RE.: Residual Error).
Fig. 2. Diagram of Semi-variance Model
2.3 Spatial Autocorrelation The spatial autocorrelation characteristics of each plot are figured out by applying GeoDa software (Table 3). It can be seen from Table 3 that the significant levels of spatial overall autocorrelations of the two mentioned lands in No. 13 Village are
Village-Level Spatial Variability of Agricultural Soil Available K
679
higher than that of Xiguan Village, which demonstrates that the constitutive property of No. 13 Village is relatively stronger, but the randomness of Xiguan Village is relatively stronger. It can be also seen that, the two mentioned lands of No. 13 Village takes high-high aggregations as the main, aggregate type, while those in Xiguan Village are comparatively well-distributed, which indicates that the differences within a same village are small, but large between villages. Comparisons are made between spatial autocorrelation of various soil available K, the d uniformity degree of soil available K between regions and between plots is fairly different. The spatial distribution of soil available K is directly related to the peasants’ fertilizing habits, types of crops and management level. When the sampling interval does not surpass the maximum range of spatial autocorrelation of available K, the sampling and analysis towards various available K are relatively reliable and can be taken as the foundation for precision fertilizing. Table 3. Spatial Autocorrelation and Its Aggregate Type
Sap.
Sig.
Moran’I
H-H
L-H
L-L
H-L
No. 9
0.02
0.1011
22
17
19
13
No. 3
0.001
0.3830
32
11
20
7
No. 7
0.005
0.1555
33
12
12
15
(Sap.:Sampling Plot ;Sig.:Significant Level ;H-H: High-high Aggregation ;L-H :Lowhigh Aggregation ;L-L :Low-low Aggregation ;H-L: High-low Aggregation).
3 Conclusion With the example of spatial variability of soil available K, through comparing the differences between averages, coefficients of variation, ratios of nugget and sill, maximum ranges, spatial autocorrelations and aggregate types of soil available K of the plots in the same village and between different villages, the differences of villagelevel spatial variability of agricultural soil nutrients are studied, which shows that, through the long-term human influences on farmland, the difference of available K between the two mentioned lands of No. 13 Village is small, while, the difference of soil available K between No. 13 Village and Xiguan Village is relatively larger, which illustrates that human activities have changed the natural differences of soil of No. 13 Village, but have not changed those of soil between different villages. In the past researches, only changes of soil nutrients in the same plot or village have been considered, while, the differences of soil nutrients between different villages had not been taken into account. The research on spatial variability degree of soil available K in the thesis and the research on spatial variability of soil available K through collecting and analyzing soil samples can effectively supply practical
680
W. Cui and J. Liu
assistance for the fertilizing of precision agriculture, and further ensure smooth agricultural production. Because the sample-taken areas are only selected plots in two villages, the number of sampling data is small; while only the character of data themselves is considered during the process of selecting models instead of the essential character of the matter, therefore, other software would be selected to solve similar problems in future researches.
References 1. Jiang, C.: Research on the Law of Spatial Variability of Soil Available K and Its Management Technology under Different Operative Mechanism. Graduate Institute of Chinese Academy of Agricultural Sciences (June 2000) 2. Haneklaus, S., Ruehling, I., Schroder, D., Schnug, E.: Studies on the Variability of Soil and Crop Fertility Parameters and Yields in DifferentLandscapes of Northern Germany. In: Stafford, J.V. (ed.) 1st European Conf. Precision Agriculture, vol. lII, pp. 785–792. BIOS Scientific Publishers Ltd., Braunschweig (1997) 3. Jin, J.Y.: Precision Agriculture and Its Application Prospects in China. Plant Nutrition and Fertilizers Science 4, 1–7 (1998) 4. Mallarino, A.P.: Spatial Variability Patterns of Phosphorus and Potassium in No-tilled Soils for Two Sampling Scales. Soil Sci. Soc. Am. J. 60, 1473–1481 (1996) 5. Hu, Z.Y., Silvia, H., Liu, Q., Cheng, K., Cao, Z.H., Ewald, S.: Small-scale spatial variability of phosphorus in a paddy soil. Communications in soil Science and Plant Analysis 34, 2791–2801 (2003) 6. Wang, X.F., Zhang, H.: Spatial Variability of Soil Organic Matter. Soils 27(2), 85–89 (1995) 7. Zhou, H.Z., Gong, Z.T., Lamp, J.: Research on Spatial Variability of Soils. Acta Pedologica Sinica 33, 232–241 (1996) 8. Li, J.M., Li, S.X.: Spatial Variability of Some Nutrients in Soil. Research on Agriculture of Arid Region 16, 58–64 (1998) 9. Hu, K.L., Li, B.G., Lin, Q.M., et al.: Characteristics of Spatial Variability of Farmland Nutrients. Agricultural Engineering Journal 15, 33–38 (1999) 10. Yang, Y.L., Tian, C.Y., Sheng, J.D., et al.: Anthropogenic-alluvial Soil Organic Matter. A Primary Exploration on Spatial Variability of Full Dose N, P and K.J. Research on Agriculture of Arid Region 20, 26–30 11. Yang, L.P., Jiang, C., Jin, J.Y.: A Primary Exploration on Precision Management towards Cotton Field Nutrients. Scientia Agricultura Sinica 33, 67–72 (2000) 12. Bai, Y.L., Jin, J.Y., Yang, L.P.: Characteristics and Management of Soil Nutrients Variability of Different Scales. In: Jin, J., Bai, Y. (eds.) Precision Agriculture and Management of Soil Nutrients, pp. 51–57. Land Publisher of China, Beijing (2001) 13. Zhang, Y.S., Lin, Q.M., Qin, Y.D.: Quantified Analysis on Spatial Variability of Soil Nutrients in Large-scale Region. Acta Agriculturae Boreali-Sinica 13, 122–128 (1998) 14. Guo, X.D., Fu, B.J., Ma, K.: Research on Characteristics of Spatial Variability of Soil Nutrients Based on GIS and Geo-statistics—Taking Zunhua City of Hebei Province as an Example. Chinese Journal of Applied Ecology 11, 557–663 (2000) 15. Huang, S.W., Jin, J.Y., Yang, L.P., et al.: Spatial Variability of County-level Grain Field Nutrients. Chinese Journal of Soil Science 33, 188–193 (2002)
Village-Level Spatial Variability of Agricultural Soil Available K
681
16. Gao, Y.M., Tong, Y.A., Hu, Z.Y., et al.: Research on the Characteristics of Spatial Variability of Village-level Agricultural Soil Nutrients in Yellow Soil Region. Acta Pedologica Sinica 37, 1–6 (2006) 17. Qi, W.H., Xie, G.D., Ding, X.Z.: Research on Soil Sampling Density of Precision Agriculture—Taking Test and Demonstration Base of Shanghai Precision Agriculture as an Example. Chinese Journal of Eco-Agriculture 11, 48–52 (2003) 18. Peng, Z.L., Ze, Y., Li, Z.Y., et al.: The Characteristics of Spatial Variability of Agricultural Soil Available K of Karst Mountainous Areas under Village-scale. Guizhou Agricultural Sciences 36(5), 81–84 (2008) 19. Yang, Y.L., Shi, X.Z., Yu, D.S., et al.: Research on Spatial Variability of Region-scale Soil Nutrients and Its Influencing Factors. Geographical Science 28, 788–792 (2008)
Study on the Management System of Farmland Intelligent Irrigation* Fanghua Li1,2, Bai Wang2, Yan Huang2, Yun Teng2, and Tijiu Cai1,** 1
Forestry College, Northeast Forestry University, Harbin Heilongjiang, P.R China 2 Heilongjiang Water Conservancy Institute, Harbin Heilongjiang, P.R China
[email protected],
[email protected] Abstract. To achieve the unification of precise irrigation management, the system integrated GSM wireless communication technology, sensor technology, computer technology, automatic monitoring, and control technology in the study process. The system included the acquisition subsystem, the intelligent decision subsystem, and automatic control subsystem, with the advantages of the GSM internet, such as wide coverage area, powerful property of antiinterference, the remote data transmission was realized. The application of Java development platform and SQL Server 2000 achieved processing of real-time data. By testing and applicating, the system improved management level of agricultural water-saving irrigation for different accumulated temperature region, planting crops, soil type in Heilongjiang Province. Keywords: Wireless data transmission, intelligent decision-making, irrigation automatic control, farmland moisture management.
1 Introduction With the increasingly intensifying water resources scarcity and imbalance between supply and demand, countries actively explore effective water-saving ways and measures in the world [1]. Irrigation agriculture is one of the biggest water consumer in the world, the optimal management of agriculture irrigation can save large amount of water resources. The water resource of China is very deficient and unevenly distributed, which is especially more severe in north regions, therefore, it is imperative to develop water-saving agriculture [2]. Intelligent decision support system used in agriculture started in mid and later period of 70’s in the 20th century, it has become mature gradually after 30 years development[3]. Irrigation automatic control mode has many advantages such as saving-water, energy and labor saving and so on, also it can eliminate adverse effects caused by human factor in the process of irrigation, improve the accuracy of operation, it is beneficial to managing scientific the irrigation process and extending the advanced technologies. The exploration and * **
Foundation project:national science and technology support project(2007BAD88B01). Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 682–690, 2011. © IFIP International Federation for Information Processing 2011
Study on the Management System of Farmland Intelligent Irrigation
683
application of irrigation decision support system and water-saving irrigation professor system not only can provide scientific decision foundation and decision scheme for scientific management and optimal allocation of water resource, improve water utilization rate, dull the supply and demand contradiction of water resource, lighten the peasants’ work intensity[4], but also promote the development of agricultural information and science technology industry, it is very important for the realization of water-saving irrigation automation. Agricultural intelligent irrigation management system is accurate irrigation technology, which synthesizes the science and technology achievement of modern agriculture irrigation, establishes the water utilization management measures of agriculture irrigation, and realizes the irrigation regulate operation as the center. The system integrates GSM ( Global System for Mobile Communication) network wireless communication technology, sensor technology, computer technology, and agriculture irrigation automatic control technology, carries on calculation, analysis and decision-making based on collected information such as soil moisture, crop drought, and meteorology factors and so on, makes out irrigation forecast and decision, determines the time and amount of water irrigation, uses the decision results to automatically control and monitor the irrigation equipments simply and rapidly, carries out timely and suitable dynamic management of intelligent automatic irrigation for the field-crops, sets up the precision amount control field intelligent irrigation management system with remote monitoring, wireless transmission, rapid diagnosis, intelligent decision, and precision control functions, realizes modernization and automation of the irrigation water management means.
2 The Basic Structure and Characteristic of the System Field intelligent irrigation management system is divided into central monitor layer and substation monitor layer according to control grade. Central monitor layer timely monitors the field situation of each area, directly sends out instructions to control executive element work by controller, plays a general monitor role, the system is managed by the central monitor layer. Substation monitor layer is placed at the controlled field, each substation has independent control system, responsible for executing the commands of central monitor layer and users, and the control of the equipments and the management of information of the station. Central monitor layer and substation monitor layer are composed of acquisition subsystem, intelligent decision subsystem, and automatic control subsystem. The system collects the information by monitoring acquisition subsystem, the information through GSM network transfers timely or real-time to the monitor central host according to users’ requirements. The monitor central host connects with the computer, which makes scientific irrigation task list by intelligent decision system, the control center host gives out irrigation commands through GSM network, automatic control subsystem automatically performs the irrigation work according to the instruction of decision support system.
684
F. Li et al.
Fig. 1. The basic structure of field intelligent irrigation management system
2.1 Acquisition Subsystem 2.1.1 The Basic Function of the System Acquisition subsystem collects the basic information of the field by all kinds of equipments, the basic information such as soil water, atmospheric temperature, atmospheric humidity, the flow of water meter in flooding and so on. Acquisition system has the GSM network as the basic support for data transmission, has the Short Message Service short message as communication carrier[5][6][7], realizes the remote wireless bidirectional data communication from one point to multi-point. Data acquisition and transmission are the basic link of the whole system decision and control, the data acquisition and transmission, which has a lot of characteristics such as the low operation cost due to low on short messages cost, the stable reliability, the higher rate of data transmission, low error code rate and so on, perform timely and real-time data acquisition, stresses the timeliness and dynamic of data monitor, provides accurate and timely information guarantee for realizing water monitor and decision of large-area field. 2.1.2 Working Principle of the System According to the requirement of area characteristic and acquisition factors, by the operation on the interface between the man and the computer, user transfers the instruction to the host in the acquisition monitor center to acquire indexes through COM port, the host of acquisition monitor center uses the GSM wireless communication equipments, sends the setting command to the terminal of data acquisition in the form of short message to go on initialization. According to the requirement of received acquisition instruction, acquisition monitor substation collects the information from the soil water sensor, temperature sensor, humidity sensor and flow sensor in the field. Monitor substation based on solar energy battery can realize timely and real-time reception and transmission of information collected, which is highly effective. Monitor substation sends the information collected to the host of acquisition monitor center in the form of short message, users can master best
Study on the Management System of Farmland Intelligent Irrigation
685
Fig. 2. Work principle flowchart of data acquisition system
and comprehensive data information of field water moisture and meteorological factors by means of the fixity of timing acquisition and the random of real-time acquisition, then stores the data to database. The work principle is shown as in Fig.2 . 2.2 Intelligent Decision Subsystem 2.2.1 Basic Function of System Intelligent decision subsystem is a total integration system, which uses database, people and computer interaction to go on multi-mode organic combination, it assists decision-maker to realize scientific decision[8]. Intelligent decision system with the meteorological data and water moisture data as the basis, using the water balance model, combines with the theoretical basis and characteristic of crop water requirement in the system knowledge base, soil properties and irrigation methods, makes reasoning computation according to some rules, makes out water saving irrigation decision for the crop planted in one area or one field, there are some irrigation reference factors, that is, making sure the crop need to irrigate or not, and when it needs to irrigate, comprehensive soil water content can reach the lower limit index suitable for crop growing or not, weather forecast and the growing stage of the crop is in the stage suitable for regulated deficit irrigation or not. It takes protective irrigation alarming mode as soil moisture content diagnose system, timely makes out accurate irrigation decision, sends out concrete operation instruction to control subsystem through the interface of interaction between man and computer. 2.2.2 Work Principle of the System Field monitor substation sends acquisition information collected by soil water sensor, temperature sensor, humidity sensor and flow sensor through GSM network to the host of monitor center, decision system takes database to receive collected information as the basis, has mode base as the support of system calculation and statistics, has the method base as the theoretical guidance of mode base calculation and statistics, has the knowledge base as the support for system theory and experience operation, makes calculation, statistics, analysis of the information, then makes out irrigation decision, makes effective and water-saving irrigation plans, realizes scientific guidance to area irrigation. The work principle is shown as in Fig.3.
686
F. Li et al.
Fig. 3. Work principle flowchart of intelligent decision subsystem
2.3 Automatic Control Subsystem 2.3.1 Basic Function of the System Field intelligent irrigation automatic control subsystem realizes scale and automation of production, raises labor productivity[9]. Control subsystem is the remote automatic irrigation monitor system of GSM wireless communication, it transmits irrigation decision scheme instruction of intelligent decision subsystem, via the host of system control to control substations by GSM network. Control substations guide the irrigation controller to open or close according to the received irrigation decision instruction and transmit irrigation water parameters collected to computer via the host of control centre, in order to achieve the aim of wireless automatic irrigation. 2.3.2 Work Principle of the System Field monitor substation sends information collected by soil water sensor, temperature sensor, humidity sensor and flow sensor by GSM network to the host of monitor center. According to mode base, method base, knowledge base and database, decision system makes out irrigation decision. Control subsystem sends irrigation decisions via GSM network to field control substation by the host of control center. Control station realizes precision irrigation control according to single control electromagnetic valve’s open or close for irrigation decision task. The work principle is shown as in Fig.4.
Fig. 4. Work principle flowchart of automatic control subsystem
Study on the Management System of Farmland Intelligent Irrigation
687
3 Database Design Database is a data storage space for storing irrigation management decision and model operation, an application software integral with the unified management of data which is massive, complex in structure, abiding and sharing as goal, meanwhile it is independent, comprehensive, universal, sharing and easy in maintenance[4][10]. The level of database structure design directly effected both the efficiency of the system and the effects of implementation[11]. Database design includes basic database, real-time database and spatial database design, it is mainly composed of graphic library, all kinds of acquisition information and attribute information of all kinds of ground objects [12][13]. In the field intelligent irrigation management system, Database Management System is used mainly to manage and maintain the data in database, makes the information storage and reading systematic, standardized and automatic, including browsing, query, renewal, adding, backup and recovery of data and so on. Database of system used the efficient relational database to combine with Windows NT/2000 and windows 9x,the characteristics of operating system were adequately used. Combinning with GSM wireless communication technology, the management system of farmland intelligent irrigation was developed by the test function and the graphic user interface of Java virtual platform, which was based on virtual instrument technology. Established database includes six functions as follows: (1) Meteorology database. It can store, query and modify the basic information of meteorology factors, such as accumulated temperature, rainfall, atmospheric temperature and relative humidity etc. (2) Crop information database. It can store, query and modify the basic information of crop growth and development, including crop growth stage, days of growth stage, the depth of effective water absorption layer of root in each growth stage, suitable water upper and lower limits in each growth stage etc. (3) Soil information database. It can store the basic soil water dynamic information of management regions, including field capacity and bulk density (The type of soil is divided into black soil, chernozem, sandy loam soil, meadow soil and saline soil and so on ), and can query and modify the basic dynamic information of soil water. (4) Irrigation regional database. It can store, query and modify the basic information of irrigation regions, including the crop, soil, equipment, picture and characters etc in the regions. (5) Irrigation elements database. It can store, query and modify the basic information of the ways and quota of irrigation. The ways of irrigation include sprinkler irrigation, drip irrigation and furrow irrigation. The quota of irrigation includes the quota of different accumulated temperature zones, different crops and different growth stages. (6) Irrigation alarming database. It includes irrigation time, irrigation quantity and alarm limit of soil water etc.
4 Application and Analysis of the System The management system of farmland intelligent irrigation was tested and applied between the year 2009 and the year 2010 in some areas of Heilongjiang Province,
688
F. Li et al.
such as Qinggang, Dumeng, Zhaodong, meanwhile, was improved and upgrade. The system used field sensors to achieve the collecting work, the measurement accuracy of soil water sensor reached ±2% (m3/m3) in 0-50% (m3/m3 reached stable after electrifying. The solar cell of power supply system had some advantages to avoid hindering collection of information with insufficient electricity under rainy weather condition, the system could run normally, such as non equalization charging, small self-discharge, without the leakage of liquid, without overflow of acid gas. By processing the information, acquisition monitor substation timely or real-time transmitted digital signal to the host of acquisition monitor center according to users’ requirements. Acquisition monitor substation contained acquisition circuit, gate circuit, ATM processor, cpu acquired soil water by sensor, transfered the analog into digital information, the remote transmission of signal to the host of acquisition monitor center could be achieved by GSM wireless module. The host of acquisition monitor center was composed of programmable array multiplexer, constant current source, isolation transmit amplifier, 12 bit A/D converter, using 32 bitARM processor, operating system was MC/OS-II, storage capacity was 2 M byte. The host of acquisition monitor center received the GSM information from acquisition monitor substation and stored monitoring data for 9 months. The host of acquisition monitor center connected with computer, the computer intelligent irrigation management software calibrated the information of database with the support of mode base, methods base, knowledge base, the processing physical quantities were displayed software interface, processing results were stored the relative database, according to the soil moisture and growth condition of plant , decision system gave out the irrigation time, irrigation water, and other relevant information after analyzing and calculating, then, sent irrigation command to the host of control center. The host of control center was composed of programmable array multiplexer, constant current source, isolation transmit amplifier, using 32 bit ARM processor, operating system was MC/OS-II. The control substation not only were acquisition terminal but also had irrigation control performance, it was composed of GSM communication module, data acquisition module, control module, power module. The control substation received the irrigation command from the host of control center, achieved irrigation tasks with irrigation controller, then, sent executed information of irrigation tasks to the host of control center. Field intelligent irrigation management system sets up design features with area characteristics, besides some functions such as irrigation decision support system, water-saving irrigation expert system and artificial intelligence irrigation management system. (1) The system has different accumulated temperature zones, different crops and different soil types in Heilongjiang province as the basic indexes, establishes the comprehensive database combined with crop growth information and soil information essential for field irrigation, satisfies actual demand for irrigation decision and guidance under the condition of different planting structures and different characteristic regions, and also can modify, newly add parameters to the database and has better expansibility. (2) The system has the GSM network and solar energy component power supply system as support, realizes timing or real-time reception and remote wireless
Study on the Management System of Farmland Intelligent Irrigation
689
transmission of various acquisition information in field. Real-time monitor of data information provides powerful guarantee for irrigation decision. (3) The system establishes water diagnosing index system which is based on field moisture content of crop physiological characteristics and soil physical characteristics, monitored by sensor, using man and computer interaction system design, The system performs the alarming function to soil water deficiency or sufficiency, implements scientific irrigation management and system monitor maintenance. (4) The system regional management design in different layers overcomes inhomogeneity in spatial distribution on small scale, implements dynamic monitor of point to point moisture content in field divisions, satisfies the hierarchy of time distribution on large scale, realizes concentration and integration of automatic irrigation in management regions, sets up precision and effective monitor and irrigation management technique modes which satisfy regional characteristics.
5 Conclusions Field intelligent irrigation system is prepared mainly on the basis of aggregating large amount of knowledge information, in connection with characteristics of the soil properties of main crops, the law of crop water requirements, and crop evapotranspiration etc a series of data indexes of soybean, sunflower and sugarbeet, etc, planted in different accumulated temperature zones in Heilongjiang province. With the support of computer, the system uses remote monitoring, realizes the monitor of soil moisture and timely monitors the amount of water consumption and requirement in field crop growth period, gives accurately values of crop water requirement, makes out irrigation decisions, and the irrigation control system carries out automatic irrigation according to decision content, which makes dispersed agriculture facilities and management areas a whole body, improves the operation efficiency of agriculture, reduces the cost of production, provides proper moisture growth environment for crops, realizes the objects of efficiency, water saving and yield increase.
References 1. Ma, J.: The Study of Accurate Irrigation and Fertilization Automatic Control System. Shanxi Agricultural University. Master Degree Dissertation, p. 5 (2005) 2. Yang, J.: The Application of Software Engineering in the Development of Agricultural Expert System. Journal of Agricultural Mechanization Research 3, 224 (2005) 3. Deng, J.: The Design and Implementation of Intelligent Decision System for Apply Fertilizer and Irrigation. Central China Normal University. Doctorate Dissertation, p. 2 (2006) 4. Cui, J., Ma, F., Zheng, Z., et al.: Study on Field Water Irrigation Management and Automatic Control System Based on GSM. Water-saving Irrigation 5, 30–31 (2005) 5. Zhang, Z., Zhao, P., Yang, Z., et al.: System of the Hydrology Information Remote Collection Based on GIS and GSM. Journal of Anhui Agricultural Science 36, 6585–6586 (2008)
690
F. Li et al.
6. Shu, Y.: Real Time Collection System of Soil Moisture Based on Short Message. Guizhou University. Master Degree Dissertation, pp. 6–7 (2006) 7. Li, N., Liu, C., Li, Y., et al.: Development of remote monitoring system for soil moisture based on 3S technology alliance. Transactions of the Chinese Society of Agricultural Engineering 26, 169–172 (2010) 8. Zhang, W., Liu, X., Gu, H.: Development of Real-time Irrigation Decision Support System Software of Irrigation Regional. Yellow River 29, 54 (2007) 9. Hao, W., Peng, X., Geng, Q., et al.: Intelligent Irrigation Control System Based on ARM. China Rural Water and Hydropower 5, 24 (2006) 10. Shang, H., Wang, Z., Chai, P.: Research and Development of Water-saving Irrigation Database and Its Management System. Research of Soil and Water Conservation 9, 97–98 (2002) 11. Yang, J.: Research on Decision Making Support System for Precise Irrigation of Wheat in the Hexi Oasis Irrigation Areas. Gansu Agricultural University. Master Degree Dissertation, pp. 42–43 (2007) 12. Sun, M., Cai, D.: Research on Irrigation Areas Water Resource Management Based on Ground Information System. Water Saving Irrigation 1, 43 (2007) 13. Chen, L., Huang, J.: Integration of Irrigation Management Model and GIS and Its Application. Journal of Irrigation and Drainage 3, 29–30 (2003)
Extracting Winter Wheat Planting Area Based on Cropping System with Remote Sensing* Xueyan Sui1, Xiaodong Zhang1, Shaokun Li2,3,**, Zhenlin Zhu1, Bo Ming2, and Xiaoqing Sun1 1
Institute of Agriculture Sustainable Development/Shandong Academy of Agriculture Sciences, Jinan 250100, China Tel.: 0531-83179362
[email protected] 2 Institute of Crop Sciences, Chinese Academy of Agriculture Sciences/ Key Laboratory of Crop Physiology and Production Ministry of Agriculture, China, Beijing 100081, China Tel.: 010-82108891
[email protected] 3 Key Laboratory of Oasis Ecology Agriculture of Xinjiang Construction Crops/ The Center of Crop High-yield Research, Shihezi 832003, China
Abstract. Winter wheat is one kind of important crop in China. It’s planting area is one key element to explain yield change. To obtain winter wheat planting area as soon as possible can provide scientific reference for our country’s making related policy. Basing on the cropping system in Shandong province, winter wheat is divided into two kinds “winter wheat sowed by machine-maize” and “people broadcast winter wheat-rice”. Using MODIS data, NDVI characters of winter wheat, garlic, greenhouse vegetable, from sowing till overwintering stage were analyzed. Together with NDVI characters of former stubble crops in middle September, extracting requirements were set up for winter wheat planting area which was sowed by machine this year. In view of the spectrum similarity between rice wheat and greenhouse vegetable from sowing stage till overwintering stage, rice wheat planting area of former year was extracted relying on the character of biomass rapid growth at jointing stage. Because of the “people broadcast winter wheat-rice” cropping system is very fixed in Shandong province, then the rice wheat planting area of former year can take the place of the rice wheat planting area this year. Two kinds of winter wheat area were merged, and tested by 284 groups of located spots data, the accuracy reached 94.01%. The result showed that it is feasible to extract winter wheat area before overwintering stage, and the time is 4 months earlier than using jointing stage NDVI. Keywords: Shandong province, winter wheat, rice wheat, area, remote sensing.
*
Fund projects: National Science & Technology Pillar Program (2007BAH12B02); Shandong Academy of Agricultural Sciences innovation program (2007YCX026); Shandong Province Science & Technology development projects (2009GG10009007). ** Corresponding author. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 691–699, 2011. © IFIP International Federation for Information Processing 2011
692
X. Sui et al.
1 Introduction Shandong Province located in the Huang-Huai Plain is the major winter wheat producing area in China. It is important to grasp winter wheat area timely and accurately for the country's food security. In the past, winter wheat area mainly relied on primary agricultural production sectors’ artificial survey and then statistics level by level. Limited resources and subjectivity led to poor data, which brought some impact to government’s decision on agriculture production. With the development of science and technology, In 1980s, remote sensing had become one tool of agriculture resource investigating and crop growth monitoring with characters of obtaining data objectively, accurately and timely [1].The key to extract winter wheat area is the data resource and the choice of phase. In 1990s, using NOAA satellite data, Maoxin Wang set up regression equation of winter wheat area and the pixel number whose NDVI (Normal Differential Vegetation Index) difference between November and October was greater than zero [2]. A large population but less land area, complex cropping system and NOAA satellite data’s lower spatial resolution resulted that the extracting accuracy of winter wheat was not high. In the 21st century, with the food strategy advancement, some regions extracted winter wheat area with TM and spot data [3, 4] which holds higher spatial resolution than NOAA data. Extracting methods included visual interpretation [5], supervised classification [6], unsupervised classification [7, 8] and pure pixel identification based on spectral library [9]. Although resolution being enhanced and data processing technology being improved have made the accuracy reach over 90%, While TM data has lower time resolution and easily influenced by weather, what’s more the cost is high, then TM data can only be used to invest winter wheat areas of small regions but not large areas. Compared with NOAA data and TM data, MODIS data has higher time and middle spatial resolution. The data sharing service has made MODIS data be used to remote sensing monitoring winter wheat more and more [10-13]. Yigang Jing set models and extracted winter wheat area with accuracy of over 91% using the NDVI changes of March, May and June [14]. Jinqiu Zou extracted area using the EVI (Enhanced Vegetation Index) difference between May and October, and the error rate was -0.04% [15]. Wenpeng Lin used NIR, RED, BLUE and ESWIR 4 bands MODIS data of October and December to class 6 kinds of surface features with the method of fuzzy ARTMAP, and the accuracy reached 80.3% [16]. The main method of comparing spectral changes in the key growth periods after sowing for extracting winter wheat area was used frequently using MODIS data [17]. Over time, the amount of information increased, and the accuracy enhanced [18-20]. The paper studied one method with high accuracy to extract winter wheat area at early periods, combining with the former stubble crops based on cropping system in Shandong province.
2 Data Acquisition Winter wheat is sowed in middle to late October in Shandong province, and tillers in early to middle December. In middle to late October, most vegetables in greenhouse are transplanted, and grow fast after rejuvenation in December, and the change of NDVI is similar with that of winter wheat, which influences the extracting accuracy of winter
Extracting Winter Wheat Planting Area Based on Cropping System
693
wheat area [16]. Shandong province is also the major producing area of garlic. It is sowed in late September to early October, the growing condition is very similar to winter wheat in winter, and thus it is another important obstacle. In order to exclude the interferences of greenhouse and garlic, discussion group located 284 spots of six kinds of surface features which included winter wheat, greenhouse vegetable, garlic, village, uncovered cotton filed, tree, in Shandong province, in late October, 2008. Downloaded MOD09Q1 data of the 284 spots of middle January, middle April, middle September, middle October and middle December, form MODIS data sharing platform-ftp://e4ftl01u.ecs.nasa.gov/.
3 Data Analysis 3.1 This Part Analyzed Surface Features’ NDVI Sequence, and Set Up Identification Conditions to Extract Winter Wheat Area Preliminarily We calculated the located spots’ NDVI and the average of middle September, middle October and middle December for six kinds of surface factures separately, and then drew line chart (fig. 1). The NDVI of cotton filed dropped as the season went on, and it was similar to village and tree. Among the three periods, the NDVI in September was the highest for winter wheat, greenhouse vegetable and garlic, at that time, winter wheat filed and garlic filed are all planted maize, and greenhouse is still planted vegetable. The gain period of garlic is in late May, which is earlier than winter wheat. After garlic the filed is planted with early mature maize, and the maize is gained in middle to late September. While winter wheat is gained in middle June, then the filed is planted with maize. Winter wheat field’s maize is at filling period, so the NDVI average 0.71 in middle September is higher than that of garlic filed. Till middle October, maize is gained, and winter wheat is sowed. Garlic is sowed in late September to early October, and the time of emergence is earlier than winter wheat. Greenhouse
0.8
winter wheat
0.7 greenhouse vegetable
0.6 NDVI
0.5
garlic
0.4 tree
0.3 0.2
cotton
0.1 village
0 September
October
December
month
Fig. 1. NDVI line chart of 6 kinds of surface features in the three periods
694
X. Sui et al.
vegetable is transplanted, and grows faster than winter wheat and garlic with good conditions. Thus greenhouse vegetable’s NDVI is higher than garlic field’s, and garlic field’s NDVI is higher than winter wheat field’s. In middle December, winter wheat steps into tillering stage, and reaches growing peak before winter. Garlic’s biomass and greenhouse vegetable’s biomass all continues to increase. So in middle December, the three kinds surface features’ NDVI are all higher than that in middle October, but lower than in middle September. According to the analysis above all, preliminary identification conditions were set up to extract winter wheat area: NDVI middle September > NDVI middle December > NDVI middle October and NDVI middle September >0.5
,
。
3.2 Test on Identification Result Used the identification conditions to extract winter wheat area, and tested the result with 284 groups of located spots’ data (table 1). Table 1. The test results checklist The number of
The number of spots
The number of spots
spots to be tested
tested to be Winter Wheat
tested to be others
winter wheat
161
138
23
cotton
24
0
24
——
31
3
28
mixed pixels mixed pixels
Surface features
greenhouse vegetable
Wrong test reasons 7 mixed pixels 16 rice wheat
garlic
11
3
8
tree
26
1
25
mixed pixel
village
31
3
28
mixed pixels
Winter wheat area extracted Located spots Ji Ning
Fig. 2. Spots of rice wheat tested to be others
Extracting Winter Wheat Planting Area Based on Cropping System
695
Winter wheat area extracted Located spots
He Ze
Fig. 3. Mixed pixel of winter wheat tested to be others
The result was listed in table 1. Among 284 located spots, 23 winter wheat spots were identified to be other surface features. 10 other surface features spots were identified to be winter wheat, which included 3 greenhouse vegetable spots, 3 garlic spots, 1 tree spot and 3 village spots. Imported the 284 groups of located spots into ENVI, and inspected one by one, then discovered that the wrong identifications can be divided into two kinds of situations: one was located spots were mixed pixels of winter wheat with greenhouse vegetable, garlic, tree or village (fig. 2); the other was rice wheat spots were identified to be other surface features (fig. 3). A large population but less land area and complex cropping system in Shandong province make the phenomenon of mixed pixels inevitable. Rice wheat was not identified, showed the identification conditions didn’t contain rice wheat information. Only extracted “winter wheat sowed by machine-maize” area, so the main task remained was to extract “people broadcast winter wheat-rice” area. 3.3 The Extraction of Rice Wheat Area Calculated 16 rice wheat spots’ average NDVI of the 3 periods, compared with common winter wheat. As fig.4, fig5 showed, in September, rice is growing in rice wheat field, and maize is growing in winter wheat field, and the NDVI of rice is lower than that of maize. In middle October rice is gained, and rice wheat is at trefoil stage which was broadcasted not long ago, whose NDVI is higher than normal wheat. What’s more rice wheat field has adequate water supply, so rice wheat growing condition is better than normal winter wheat till middle December. During the analyzing progress, narrowing NDVI threshold of the 3 periods was tried to extract rice wheat area, but failed for the NDVI similarity between rice wheat and greenhouse vegetable. Winter wheat goes into overwintering stage In January, and jointing stage in April, growing condition is influenced by air temperature; rice wheat and normal winter wheat are similar to each other. So it is possible to extract all winter wheat area depending on NDVI fast increasing over the period from middle January to middle April, that is to say, from overwintering stage to jointing stag. This method is popular now.
696
X. Sui et al.
0.8 0.7 0.6
rice wheat
NDVI
0.5 normal winter wheat
0.4 0.3
greenhouse vegetable
0.2 0.1 0 September
October
December
month Fig. 4. NDVI line chart of rice wheat, normal winter wheat, greenhouse vegetable in the three periods
Fig. 5. Winter wheat sowed by machine and people broadcast winter rice wheat in middle to late October
Affected by geographical environment and climate, “people broadcast winter wheat-rice” is one common cropping system in Yellow River Basin and the Southern Four Lakes rim, so the rice wheat field is relatively fixed. According to Statistical Yearbook, vector graph of rice planting region was set up in Shandong province with the help of ARCVIEW. The normal winter wheat area which has been extracted with data of middle September, middle October, middle December was named S1. The rice wheat area extracting conditions were set up, (NDVI middle April - NDVI middle January) > 0.19, NDVI middle January > 0.3, and NDVI middle September >0.5, using the data of 2008. In ENVI, S1 was masked, and among rice wheat field vector region, rice wheat area was extracted of 2007-2008, which were named S2 (fig. 6). The 16 rice wheat located spots were all extracted successfully.
Extracting Winter Wheat Planting Area Based on Cropping System
697
Fig. 6. The extracted area of rice wheat of 2007-2008
3.4 The Extraction of all Winter Wheat The two parts S1 and S2 of winter wheat were merged, which contained the normal winter wheat area extracted with the data of September, October, and December, 2008,
Fig. 7. The extracted area of winter wheat of Shandong province of 2008-2009
698
X. Sui et al.
and the rice wheat area extracted with the data of January, April, and September, 2008. The total winter wheat area of 2008-2009 was 4480666.67 hm2, tested by the located spots’ data the accuracy was 94.01%(fig. 7).
4 Result and Discussion Winter wheat in Shandong province can be divided into two kinds from a cropping system point of view, “mechanical sowing winter wheat-maize” and “people broadcast winter wheat-rice”. The paper studied NDVI characters of winter wheat, garlic and greenhouse vegetable from seeding time October to overwintering stage December. The study showed that 3 kinds of surface features’ NDVI changing tendencies are similar to each other, and it is difficult to extract winter wheat area. Together with the difference of the previous crop’s NDVI, the normal winter wheat of “mechanical sowing winter wheat-maize” was extracted. But the rice wheat NDVI of “people broadcast winter wheat-rice” is similar to greenhouse vegetable NDVI, so it is impossible to extract rice wheat area just depending on the change of NDVI. Based on the special growing environment of rice wheat, rice plant region vector in Shandong province was set up. Among this region, rice wheat area of the previous year was extracted using the fast NDVI increase at jointing stage than overwintering stage. The two parts winter wheat area were added up, and got the total area in late December. This study used middle spatial resolution and high time resolution MODIS data to settle the interference of garlic and greenhouse vegetable to winter wheat area extracting in Shandong province. The extracting time was 4 months in advance than using jointing stage NDVI character[21,22], and the accuracy reached 94.01%, so the method can meet the demand of large area. Natural environment decides cropping system, while the improvement of agricultural production technology, and many other economical factors’ all influence the reform of cropping system. So using the previous year rice wheat area to take the place of the current year will affect the accuracy of total winter wheat area. In future research, rice field’s spectrum should be studied, and assisted with NDVI change character of rice wheat from seeding to overwintering stage to extract rice wheat area, in order to enhance the total winter wheat area extraction accuracy.
References 1. Mei, A., Peng, W., Qin, Q., et al.: Introduction to Remote Sensing. Higher Education Press, Beijing (2001) 2. Wang, M., Pei, Z., Wu, Q., et al.: Winter wheat sown area estimation using NOAA AVHRR data. Transactions of the CSAE 14, 84–88 (1998) 3. Murakami, T., Ogawa, S., Ishitsuka, M., et al.: Crop discrim-ination with multitemporal SPOT/HRV data in the Saga Plains, Japan. Int J. Remote Sens. 22, 1335–1348 (2001) 4. Qi, L., Liu, L., Zhao, C., et al.: Selection of optimum periods for extracting winter wheat based on multi-temporal remote sensing images. Remote Sensing Technology and Application 23, 154–160 (2008)
Extracting Winter Wheat Planting Area Based on Cropping System
699
5. Gu, X., Pan, Y., Zhu, X., et al.: Consistency study between MODIS and TM on winter wheat plant area monitoring——a Case in small area. Journal of Remote Sensing 11, 350–357 (2007) 6. Feng, M., Yang, W., Zhang, D., et al.: Monitoring planting area and growth situation of irrigation-land and dry-land winter wheat based on TM and MODIS data. Transactions of the Chinese Society of Agricultural Engineering 25, 103–109 (2009) 7. Huang, X., He, W., Zhang, Y., et al.: Monitoring on wheat area using TM in some areas of Jiangsu province. Jiangsu Agricultural Sciences, 85–87 (2003) 8. Li, W., Li, H., Wang, J., et al.: A study on classification and monitoring of winter wheat growth status by Landsat/TM image. Journal of Triticeae Crops 30, 92–95 (2010) 9. Chen, S.: Remote sensing method of pure crop pixel identification and planting area estimation based on spectral library. Chinese Academy of Science, Beijing (2005) 10. Yang, X., Zhang, X., Jiang, D.: Extraction of multi-crop planting areas from MODIS data. Resources Science 26, 17–22 (2004) 11. Wu, Y., Wang, Y., Zhang, J., et al.: Linear mixture modeling applied to remote sensing monitoring of winter wheat areas. Transactions of the Chinese Society of Agricultural Engineering 24, 136–140 (2009) 12. Yan, F., Wang, Y., Wu, J., et al.: Extracting winter wheat area using temporal sequence of Ts-EVI. Transactions of the CSAE 25, 135–140 (2009) 13. Qiao, H., Zhang, H., Cheng, D.: Application of EOS/MODIS-NDVI at different time sequences on monitoring winter wheat acreage in Henan Province. Journal of Anhui Agricultural Sciences 36, 11940–11941 (2008) 14. Jing, Y.: Study on extracting winter wheat area using EOS/MODIS data. Shanxi Journal of Agricultural Sciences (2), 95–98 (2008) 15. Zou, J., Chen, Y., Uchida, S., et al.: Method for extracting winter wheat area using Terra/MODIS data and its accuracy analysis. Transactions of the Chinese Society of Agricultural Engineering 23(11), 195–200 (2007) 16. Lin, W.: Study on crop information extraction based on MODIS spectral analysis. Chinese Academy of Science, Beijing (2006) 17. Zhang, M., Zhou, Q., Chen, Z., et al.: Crop acreage change detection based on phenology model. Transactions of the CSAE 22, 139–144 (2006) 18. Wang, Y., Shen, R., Tian, G.: Study on Extracting Winter Wheat Planting Area Based on MODIS Data by Spectral Mutation. Nei Menggu Weather, 18–21 (2009) 19. Wang, Y., Shen, R.: Study on the Planting Area Extraction of Winter Wheat Based on MODIS Data. Journal of Anhui Agricultural Sciences 37, 16694–16696 (2009) 20. Xu, W., Zhang, G., Fan, J., et al.: Remote sensing monitoring of winter wheat areas using MODIS data. Transactions of the CSAE 23, 144–148, 196 (2007) 21. Qin, Y., Zhao, G., Jiang, S., et al.: Winter wheat yield estimation based on high and moderate resolution remote sensing data at county level. Transactions of the CSAE 25, 118–123 (2009) 22. Chen, J., Liu, H., Huang, Y., et al.: Analysis on the Multi-temporal Features of MODIS/NDVI in the Growing Period of Winter Wheat and Its Application in Ground-object Identification. Journal of Anhui Agricultural Sciences 38, 3641–3643, 3667 (2010)
Study on the Rainfall Interpolation Algorithm of Distributed Hydrological Model Based on RS Xiaoxia Yang, Yong Liang, and Song Jia School of Information Science and Engineering, Shandong Agricultural University, Taian, Shandong Province, P. R. China, 271018
[email protected] Abstract. Distributed hydrological model can be divided into two parts, as runoff and the convergence. Runoff calculation is the basis of distributed hydrological model, its results will determine the accuracy of the simulation results directly. Rainfall is an important input of runoff calculation, thus its accuracy has special significance to runoff calculation. For most small watershed, now mainly rely on rainfall stations to obtain rainfall information. In this case, the most effective approach is base on space correlativity principle, use the interpolation algorithm to obtain distributed rainfall data. Take XueYe reservoir 2000-2008 year of reality measures data as an example, we compared the apply condition of several kinds algorithm such as co-kriging interpolation, kriging interpolation, reverse-weighted interpolation, and so on, prove that co-kriging interpolation is most fit XueYe reservoir . Keywords: Distributed hydrological model, Association Kriging interpolation, Kriging interpolation, Reverse-weighted interpolation.
1 Introduction Distributed hydrological model has been a hot research field for nearly 20 years [1], what can be divided into two parts, as runoff and the convergence. Runoff calculation is the basis of distributed hydrological model, its results will determine the accuracy of the simulation results directly. Rainfall is an important input of runoff calculation, its spatial distribution characteristics is the main control factors of runoff and a series of other hydrological problems. The effective method to get accurate rainfall distribution characteristics is to set density rainfall station observation network [2],but now most small watershed has limited quantity of rainfall station and the distribution of rainfall stations are always not reasonable, so the data from these stations often cannot meet the requirements. Therefore, the spatial interpolation method according to the acquired data become a research hotspot. There are several interpolation methods [3] [4] [5] [6] [7], different method has different results, and no one is common interpolation method for optimum interpolation method [8] [9].Taking XueYe reservoir 2000-2008 year of reality measured data as an example the paper contrast cokriging interpolation, kriging interpolation, reverse-weighted interpolation and thiessen polygon interpolation, then analyses the interpolation method fit XueYe reservoir. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 700–705, 2011. © IFIP International Federation for Information Processing 2011
Study on the Rainfall Interpolation Algorithm of Distributed Hydrological Model
701
Fig. 1. 11 rainfall stations’s scattergram of XueYe reservoir
XueYe reservoir is one of the large reservoir in ShanDong province, founded in 1959.XueYe reservoir has a total capacity of 2.21 billion cubic meters, water area of 1.8 million acres. It located in 117.54 ~ 117.64 degrees east longitude and latitude between 36.39~36.46 degrees, belong to temperate continental monsoon climate, the four seasons, the annual average temperature is between 11.0- 13.0 degrees Celsius, precipitation 760.9 mm. Figure 1 is about 11 rainfall stations’s scattergram of XueYe reservoir.
2 Rainfall Space Interpolation Method First, compare the relationship between the XueYe 2000-2008 reality measures rainfall data and longitude, latitude, elevation, found that rainfall data with latitude, longitude linear correlation is not obvious, the correlation coefficient R2 are -0.52 and -0.47.But the rainfall data with elevation is obvious, R2=0.6021. 2.1 Thiessen Polygon Interpolation Thiessen polygon interpolation is one of the most simple partial interpolation algorithm, put forward by the Dutch climatologists A.H. Thiessen. Its main idea is adjoin all rainfall stations by triangle, for each side of the triangle do vertical bisectrix, each station around several vertical bisectrix form a polygon, and the polygon contain only one station, the polygon called Thiessen polygon. Thus we can express the data of all the points in the polygon by the station’s reality measures data. Thiessen polygon interpolation is simple. In the case that there are enough rainfall stations, Thiessen polygon interpolation can be good at approximating the actual value. It’s shortcoming is the mutation on the boundary. It can’t fit the space characteristics that rainfall is gradually changing. Meanwhile Thiessen polygon interpolation neglect the influence of elevation, not suitable for data interpolation of XueYe reservoir.
702
X. Yang, Y. Liang, and S. Jia
2.2 Reverse-Weighted Interpolation Reverse-weighted interpolation is one of the space geometric interpolation method. When we use the sampling points estimate the estimation point, often farther to the estimate point smaller the influence is, in other words farther sampling point has smaller weight. Thus, the value of eatimate point Z(x0) can be fitting by all the surrounding point’s linear weighted: n
1 Z ( xi ) P i =1 ( Di )
Z ( x0 ) = ∑
n
1
∑ ( Di) i =1
(1)
P .
Di is the distance from the i sampling point to eatimate point, Z(xi) is the value of i sampling point, index p is the power of distance, used to control the weight’s change speed with distance. Different value of p will influence the result of interpolation, the bigger the value of p the smaller weight of the far distance point. p can values 1,2,3.When p=2 is called the inverse distance square interpolation method. In this paper the value of p is 2.The strong point of reverse-weighted interpolation is it can adjust the result of spatial interpolation by weight, but the shortcoming is it also neglect the influence of elevation. 2.3 Kriging Interpolation Kriging interpolation is regarded as one of the main method of geological statistics, is raised by South Africa scientist D.G. Krige in 1951 and named by his name. Kriging interpolation fully absorb the idea of space statistics, think any space of continuous change attributes is very irregular, can’t imitation by simple smooth mathematical functions, but can described by random surface function. From the angle of interpolation Kriging interpolation is one of the method that is linear interpolation optimal, unbiased estimation for directional distribution data. Kriging interpolation is defined as: n
Z( x 0 ) = ∑ λi Z ( xi )
(2)
.
i =1
Z(xi) is the observed value of i point, Z(x0) is the value of estimation point,
λi
is the
power ,and we have n
∑λ i =1
i
=1
.
(3)
The linchpin of kriging interpolation is to fix the value of power λi ,it should make the value of Z(x0) unbiased estimation, that is less than any variance of linear combination observations of the variance. In the ordinary use half variance as a basis to fix the value of the power. So at the first we should fix the model of half variance, the common model of half variance are nugget, spherical surface, index, Gauss, power and linear model. In the paper we adopt spherical model.
Study on the Rainfall Interpolation Algorithm of Distributed Hydrological Model
703
2.4 Co-kriging Interpolation The theory of co-kriging interpolation is similar with kriging interpolation. Co-kriging interpolation optimal the estimation through the way that consider moer than one variables and the relation between the variables. Use co-kriging interpolation,we can use the correlation between the variables, estimate the value of one or more variables, improve the accuracy and rationality of estimation. Co-kriging interpolation introduce the crossover-mutation function, that is the function that the correlation of two different variables change with the distance. The crossover-mutation function is defined as : rij(h)=1/2*E{[Zi(x+h)- Zi(x)][ Zji(x+h)- Zj(x)]}
(4)
Within XueYe reservoir the elevation is knowed ererywhere and stable,we introduce the elevation as a factor into co-kriging interpolation.
3 Analysis the Result of Rainfall Interpolation 3.1 Method of Calibration Adopt cross validation methods to analysis the result of interpolation: randomly extract 2 rainfall stations as docimastic station among all the 11 rainfall stations of XueYe reservoir, remove the data of the docimastic station, use data of the rest 9 rainfall stations to estimate the value of docimastic staion, and compute the error. Use mean absolute error(MAE) and root mean squared interpolation error(RMSIE) as the evaluation criterion to evaluate the result of interpolation. MAE is used to evaluate the error range, RMSIE can reflect valuations sensitivity and maximum effect of interpolation. n
MAE = ∑ Z ' ( xi ) − Z ( xi ) n
(5)
.
i =1
RMSIE =
n
∑[ Z ' ( x ) − Z ( x )] i
2
i
i =1
Table 1. The result of reverse-weighted interpolation
n
.
(6)
704
X. Yang, Y. Liang, and S. Jia Table 2. The result of kriging interpolation
Table 3. The result of co-kriging interpolation
Zi is the actual observed value of the i point, Z’i is the estimation value of it, n is the amount of docimastic station. Prepeocess the data will make the data tends to normal distribution and improve the accuracy of estimation. In this paper we adopt square root transformation to improve the accuracy of estimation. 3.2 Analysis the Result of Interpolation Paper use square root transformation, adopt reverse-weighted interpolation, kriging interpolation, co-kriging interpolation, get 6 results of interpolation. Compare the 6 results, find the best interpolation fits to XueYe reservoir. Analysised years of data we find that: most month have better interpolation accuracy after square root transformation. Compare with original data co-kriging interpolation after square root transformation has better interpolation accuracy in month 1,2,3,4,5,8,9,10,11,12.Compare with original data kriging interpolation after square root transformation has better interpolation accuracy in month 1,2,4,5,6,7,10,11,12. Compare with original data reverse-weighted interpolation after square root interpolatio has better interpolation accuracy in month 1,3,4,5,6,7,8,10,11,12. Compare with the other two interpolations after square root transformation cokriging interpolation has better interpolation accuracy in month 1, 3, 4, 5, 8, 9, 10, 12. Original co-kriging interpolation has better interpolation accuracy in month 7. After square root transformation kriging interpolation has better interpolation accuracy in month 2,6,11. The MAE of co-kriging interpolation after square root transformation is 3.33, RMSIE is 4.29; The MAE of original data co-kriging interpolation is 3.42,RMSIE is 4.38; The MAE of kriging interpolation after square root transformation is 3.41, RMSIE is 4.38; The MAE of original data kriging interpolation is 3.47, RMSIE is
Study on the Rainfall Interpolation Algorithm of Distributed Hydrological Model
705
4.45; The MAE of reverse-weighted interpolation after square root transformation is 3.53, RMSIE is 4.61; The MAE of original data reverse-weighted interpolation is 3.64, RMSIE is 4.63. Through the analysis of data we find that, co-kriging interpolation after square root transformation has a higher precision than the other interpolation, is more suitable for the actual situation XueYe reservoir.
4 Conclusion Analysised years of data we find that:the rainfall of XueYe has relatively close relationship with elevation, so the interpolation irrelevant with elevation is not fit to this area. Because after square root transformation the data will tends to normal distribution, so before interpolation we adopt square root transformation, experimental data shows that after square root transformation most month have better result. After analysis several common interpolation, we get the following conclusion: co-kriging interpolation after square root transformation is fit to XueYe reservoir.
Acknowledgements This study has been funded by Office of ShanDong Government Flood Control and Drought Relief Headquarters. The project name is the research and development of the system of optimal scheduling about flood resource. It is supported by Shandong Agricultural University. Sincerely thanks are also due to XueYe reservoir for providing the data for this study.
References 1. Goovaerts, P.G.: Approaches for incorporating elecation into the spatital interpolation of rainfall. Journal of Hydrology (228), 1113–1291 (2000) 2. Band, C.: Forest ecosystem process at the warter scale:Basis for distributed simulation. Ecol. (56), 171–196 (1991) 3. Christopher, D., Waync, P.G., George, H.T., Gregory, L.J., Phillip, P.: A knowledge-based approach to the statistical mapping of climate. Climate Research (22), 99–113 (2002) 4. Dabid, T.P., Daniel, W.M., Ian, A.: A comparison of two statistical methods for spatital interpolation of Canadian monthly mean climata data. Agriculutural and Forest Meteotology (101), 81–94 (2000) 5. Dirks, K.N., Stow, C.D.: Highresolution studies of rainfall on Norfolk Island, Part II:interpolation of rainfall data. J. Hudrol. (208), 187–193 (1998) 6. Bartier, P.M., Keller, C.P.: Multivate interpolation to incorporate thernatic surface data using inverse distance weighting. Computer & Geoscience (22), 795–799 (1996) 7. Huiyi, Z., Shaofeng, J.: Uncertainly in the spatial interpolation of rainfall data. Progress in Geography (23), 34–41 (2004) 8. Houghton, J., Meira, L.G., Callander, B.A.: Change 1995: the Science of Climate Change. Journal of Hydrology (1996)
Study on Vegetable Field Evaluation Index System for Non-Point Source Pollution of Dagu River Basin Jinheng Zhang1, Junqiang Wang2, Yongliang Lv1, Jianting Liu1, Dapeng Li1, Zhenxuan Yao1, Xi Jiang1, and Ying Liu1 1 Institute of Eco-environment & Agriculture Information, School of Environment and Safety Engineering, Qingdao University of Science and Technology, Qingdao, Shandong 266042, China 2 Qingdao Station for Popularizing Agricultural Techniques, Qingdao, Shandong 266000, China
Abstract. This research combined data from soil, terrain and meteorological surveys, with quantitative data from soil and water field studies to determine the leaching processes of nitrogen and phosphorus contaminants from field vegetation into the Dagu River. Using the Gray correlation analysis method, precipitation and irrigation were established as the major driving force behind leaching losses from cultivated land. The type of fertilizers used, and the chemical and physical properties (porosity and bulk density) of soils in the region were set as parent sequences. COD values of nitrogen and phosphorus concentration were calculated from soil and water samples. Also, the depth of core samples was factored into the analysis. The result showed that the supply of soil nitrogen and phosphorus, which were mainly derived from fertilizers, were not generally high levels in Qingdao. According to fuzzy association degree, this investgation confirmed main influencing factors which affected COD, total nitrogen, total phosphorus of soil and groundwater, ammonia nitrogen and nitrate nitrogen of groundwater respectively. And according to these factors the initial vegetable field evaluation index system for non-point source pollution of Dagu River basin was constructed. Keywords: Gray correlation analysis, Nitrogen, Phosphorus, COD, Precipitation, Soil physical properties, Soil type, Index system.
1 Introduction Excessive nitrogen and phosphorus values used in fertilization will significantly decreases the economic benefits of enhanced crop production, as well as increases the risk of nitrogen and phosphorus contaminants in water reserves [1-3]. Approximately, 60% of water pollution is caused by non-point source NPS) pollution in America [4]. In the Northern Australia, NPS pollution flowing into water reserves is also the major source of nitrogen contamination [5]. In Denmark, 94% of the nitrogen load and 52% of the phosphorus load in 270 rivers is a result of non-point source processes [6]. The same effects can be seen in the Netherlands, where 60% or the total nitrogen contaminants and 40-50% of the total phosphorus contaminants are caused by NPS pollution [7]. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 706–715, 2011. © IFIP International Federation for Information Processing 2011
Study on Vegetable Field Evaluation Index System for NPS Pollution
707
In recent years, as agricultural development mirrors population growth in China, NPS pollution has become the main source of water pollution. Currently, China contains 1600 million hm2 of cultivated land, which amounts to 35 of the agricultural land on the planet [8]. Statistics measuring the amount of fertilizers, livestock and poultry manure and the cultivated land area from 2000 to 2006 in Qingdao were investigated. Subsequently, an index was established to evaluate the nitrogen and phosphorous content in livestock manure as NPS pollutants [9]. A Grey Correlation study to evaluate water pollution from vegetable fields in the Dagu River Basin of Qingdao[10]. Over the years, the Dagu River has become increasingly more important to the national economy and to the lives of local residents. However, as the region surrounding the Dagu River Basin is developed, pollution of the river has become increasing obvious. It is the fertilizers and pesticides that have aided the agricultural development that threaten the water reserves. The numerous and excessive applications of fertilizers and pesticides cause soil erosion and serious water pollution from increasing nitrogen and phosphorus concentrations. This study integrates data from soil, terrain and meteorological surveys of Dagu River basin, with quantitative analysis studies of thousands of soil samples to construct an indexing system of NPS pollutants entering the Dagu River. The Dagu River is located in the western Shandong Peninsula, between 120 ° 03 '~ 120 ° 25'E and 36 ° 10 '~ 37 ° 12'N and covers an area of 4631.3 km2 in Qingdao. The river runs through five districts in Qingdao, Laixi City, Pingdu, Jiaozhou, Chengyang, Jimo (see Figure 1). The annual precipitation in the region is about 685.3 mm and an annual runoff is about 6.311×10 m3 [11].
%
2 Materials and Methods 2.1 Sample Collection For this study, soil and water samples were collect in December of 2009 from cultivated land throughout the region (Figure 1). The samples of vegetable field soil were divided into two categories, surface soil (0-20 cm) and submerged soils (80-100 cm). Additionally, surface water and ground water samples were collect from the same areas. GPS was used to map the latitude and longitude of sample areas. All samples were analyzed in the laboratory, using quantitative methods. 2.2 Sample Analysis To determine the COD of water samples, they were heated in a strongly acid solution, containing potassium dichromate. A Euro Tech ET3150B multiple digester and an ET1151M COD Monitor were used for characterization. K2S2O8 digestion molybdenum blue colorimetry was used alongside the COD digester to analyze the total concentration of phosphorus in samples. Nitrite has a visible absorption at 220nm, so a 752SP UV was used to measure its concentration in aqueous samples. Nessler's reagent was added to pretreated samples, and, then, a spectrometer was used to measure their absorbance at 420 nm. Also, oxidation-ultraviolet spectrophotometry was used to
708
J. Zhang et al.
℃
determine the total nitrogen content of samples, after the addition of potassium sulpersulthephate at 120~124 . Soil samples were first treated with alkaline potassium persulfate. Then, UV spectrophotometry was used to measure each solutions absorbance at 220nm, 275nm and 700nm to measure their total nitrogen and phosphorus content. Potassium dichromate was added to soil samples before COD analysis.
Fig. 1. Soil samples distribute
3 Results and Analysis 3.1 The Main Source and Loss of Non-point Source Pollution Fertilizer, irrigation, and atmospheric deposition are the main sources of nitrogen and phosphorus in cultivated land within the basin. Among these, fertilizers are the most abundant source of the elements. There was a large quantity of NO3--N in samples, as it is in not readily absorbed by soil colloid [12]. Rainfall and irrigation contributed to the loss of accumulated NO3--N, via surface run-off (procedure C in figure 2) and/or through leaching (procedure A and B in figure 2) [13],[14]. 3.2 The Substance Basis of Fertilizer Leaching and Driving Force of Vegetable Field Soil Nitrogen and Phosphorus A classification system was developed from the analysis of more than 7000 soil samples, which were classified on the bases of total nutrient content. Results were classified into six categories based on nitrogen supply level from vegetables. The data is shown in table 1. We analyzed more than 1000 soil samples and classified the soluble phosphorus content. Results showed that very high soluble phosphorus supply level of vegetable land was only 7.83%, but general and lower supply level were 76.99% (Table 2).
Study on Vegetable Field Evaluation Index System for NPS Pollution
709
Table 1. Classification total nitrogen in soil (Data from Qingdao soil, edited by the Soil and Fertilizer Workstation of Qingdao City) field supply level high extremely high
Upland area acre % —
—
Vegetable land acre % ——
—
5938
0.07
10174
3.16
general
1060672
12.14
211928
65.78
lower
4830122
55.31
100042
31.06
low extremely low
2676373 160226
30.65 1.83
— —
— —
Table 2. Classification soluble phosphorus in soil (Datas from edited by the Soil and Fertilizer Workstation of Qingdao City) field Supply level
Upland area
Vegetable land
high higher general
acre — — 277110
% — — 3.17
acre — 25207 147990
% — 7.83 45.93
lower
1797677
20.58
129624
40.24
low
3419031
39.15
19323
6
extremely low
3239513
37.1
—
—
From the above statistics, we found that nitrogen and phosphorus supply level in Qingdao vegetable land were not very high. Additionally, fertilizer was the main source of nitrogen and phosphorus. Subsequently, it was reasonable to study fertilizer as a basic source of fertilizer leaching. Meteorological precipitation data combined with crop irrigation were overwhelmingly contributed to the leaching loss of nitrogen and phosphorus from soil. 3.3 Determine Leaching Loss Factors of Nitrogen and Phosphorus according to Grey Relational Analysis Types of fertilizer and soils, the physical properties of different soils (porosity and bulk density), precipitation data, soil nitrogen and phosphorus content, and soil COD were indexes in the analysis of leaching nitrogen and phosphorous. Gray correlation was used to analyze the effect of these indexes on the contaminant concentrations in the local water reserve. An index system based on well-correlated factors related to nitrogen and phosphorous loss from agricultural soil was established. The factors are
710
J. Zhang et al.
listed below, in decreasing order: ammonia content, nitrate content, total nitrogen of groundwater, soil type, soil porosity (from shallow to deep), soil bulk density (from shallow to deep), precipitation, soil COD and soil total nitrogen content. Fertilizer plays the most significant role in affecting the quality of groundwater, followed by soil physical properties and soil chemical properties. The order of factors affecting total phosphorus content of groundwater was soil type, soil phosphorus content, soil COD, phosphate fertilizer of per hectare, soil porosity (from shallow to deep), soil bulk density (from shallow to deep) and precipitation, respectively. Phosphorus leaching from soil was the main source of phosphorus in groundwater. 3.4 Single Factor Assessment Index of Non-point Source Pollution 3.4.1 Fertilizer Grade of Nitrogen and Phosphorus Leaching Loss Compound fertilizer, with similar nitrogen and phosphorous contents commonly used fertilizers, was applied to survey region (total nitrogen content 15% according to N, total phosphorus content 15% according to P2O5). The result from formula (1, 2) gave a mean value of fertilizer in survey region of 150.7 kg/hm2. The standard deviation was 139.6. Mean value: n
x=
∑x
i
i =1
(1)
n
Standard Deviation: n
σ=
∑(x
i
− x) 2
(2)
i =1
n
Probability Density:
f ( x) =
1 2π ⋅ σ
e
−
( x−x)2 2σ 2
(3)
According to probability density calculations (figure 2), the second center should be considered as the average nitrogen concentration in fertilizer per hectare in every season. The standard deviation was divided into five intervals: 0~80, 80~220, 220~360, 360~500 and >500 kg·hm-2·season-1 (see table 4). An advantage of this division was that similar values can be drawn in same level, which reduces the risk that similar dispersed within the data. 3.4.2 The Precipitation Grade of Nitrogen and Phosphorus Leaching Loss When long-term soil leaching occurs, the optimum conditions were as follows: the effects of the sum of precipitation and irrigation are more than the effects of the sum of run-off, evaporation and good soil infiltration. Conditions of short-term soil leaching
Study on Vegetable Field Evaluation Index System for NPS Pollution
711
mirrored large amount of precipitation or irrigation [15]. In the study area, precipitation is heaviest from June to August, about 500mm, which accounts for more than 50% of the annual precipitation. In the summer irrigation was lessened, due to the increase precipitation. In this analysis, 200mm precipitation was used as the lower limit and was divided into 5 levels in 100 mm intervals. Short-term soil loss reached measurable level when single precipitation/irrigation events exceeded 50mm (see the table 5). Table 3. Sequence of gray correlation Table 3a
Ammonia Nitrogen
Groundwater Nitrate Nitrogen
Total Nitrogen
Nitrogen fertilizer
1
1
1
Soil type
2
2
2
shallow layer
3
3
3
medium layer
4
4
4
deep layer
5
5
5
Soil porosity
Soilbulk density shallow layer
6
6
6
medium layer
7
7
7
deep layer
8
8
8
Precipitation
9
9
9
COD of soil
10
10
10
Soil total nitrogen
11
11
11
Table 3b.
TP of groundwater soil type Soil total phosphorus COD of soil Nitrogen fertilizer Porosity (deep layer) Porosity (medium layer) Porosity (shallow layer) bulk density (shallow layer) bulk density (deep layer) bulk density (medium layer) Precipitation
1 2 3 4 5 6 7 8 9 10 11
712
J. Zhang et al.
Fig. 2. Probability density function of fertilization Table 4. Fertilizer grade of nitrogen and phosphorus leaching Compound fertilizer -2 -1 (kg.hm .season )
Grade
Description
500
F5
The light leaching loss will be happened. The degree will increase according to rate of fertilizer and water. Middle leaching loss level. But heavy leaching loss can be happened. The rate of fertilizer is more than average rate. (Super) heavy leaching loss level.
Table 5. Driving factor grade of nitrogen and phosphorus fertilizer leaching Precipitation (mm. season-1)
grade
500
W5
description Shortage of precipitation. The general leaching loss won’t be happened. Precipitation met the requirements of leaching loss. Light leaching loss may be happened. Precipitation met the requirements of leaching loss. Light leaching loss may be happened. Precipitation was adequate, and moderate leaching loss may be happened. Precipitation was extremely adequate, and heavy leaching loss may be happened.
Study on Vegetable Field Evaluation Index System for NPS Pollution
713
3.4.3 The Soil Type Grade of Nitrogen and Phosphorus Leaching Loss Soil type played an important role in leaching loss. Under similar conditions, clay soil showed minimum nitrogen and phosphorus leaching loss, whereas, the largest nitrogen and phosphorus leaching loss was from sand soil. Medium loss was observed in loam soil [16],[17]. The soils of Dagu River, which can be used to cultivate vegetables, are brown earth, aquic brown earth, brown paddy soil, cinnamon soil, eluvial cinnamon soil, developed cinnamon soil, lime concretion black soil, and fluvo-Aquic soil. Nitrogen and phosphorus leaching loss was divided into three categories, according to an analysis of the physical property of the soils, including total soil porosity and soil bulk density (see table 6). Table 6. The soil type grade of nitrogen and phosphorus leaching loss Soil type
grade
description
brown earth
-
It widely distributes in the hill, valley and the front slope of mountain. The land is thick, and there is clayey layer generally. It played reduced role on nitrogen and phosphorus leaching loss. High degree of maturation, viscous soil. It played reduced role on nitrogen and phosphorus leaching loss Thin layer, rough texture, high impurity content. Lower capability of moisture and fertilizer conservation. An enhanced nitrogen and phosphorus leaching loss type. The process of sticky soil obviously, soil deep. Higher capability of moisture and fertilizer conservation. It played reduced role on nitrogen and phosphorus leaching loss. Distribution in the slope and valley. Thick layer. Higher capability of moisture and fertilizer conservation. It played reduced role on nitrogen and phosphorus leaching loss.
aquic brown earth brown paddy soil cinnamon soil Eluvial cinnamon soil developed cinnamon soil lime concretion black soil, fluvo-Aquic soil
- + - - + - /
Thin layer. Gravel. An enhanced nitrogen and phosphorus leaching loss type. Thick layer. Sticky. It played reduced role on nitrogen and phosphorus leaching loss. Different physical and chemical properties and different degree of reposado. The soil type grade of nitrogen and phosphorus leaching loss was unclear.
Fig. 3. The Soil Body Configuration of nitrogen and phosphorus leaching loss
714
J. Zhang et al.
3.4.4 The Soil Body Configuration Grade of Nitrogen and Phosphorus Leaching Loss According to results of table 5, when sandy soil is present in the upper layer, leaching will occur more readily. On the contrary, when clay soil is present, leaching will be respectively lower. Consequently, we classified the nitrogen and phosphorus leaching loss according to the soil composition. Areas of sand-layered and thin-layered soil were classified as enhanced nitrogen and phosphorus leaching loss types. Clay layered, intercalated clay layered, Mengyu type, and Mengyin type soils were as classified as reduced nitrogen and phosphorus leaching loss types.
4 Conclusions Fertilizer was as the main source of nitrogen and phosphorus NPS pollution. Precipitation was the most significant driving force of leaching loss of nitrogen and phosphorous in regional soils. The application of fertilizer is the most significant contributor to nitrogen content in groundwater, followed by the physical and chemical properties of regional soils. The results showed that the most influential factors determining the total nitrogen content of groundwater was the soil type, chemical properties of the soil, phosphorus content, the physical properties of the soil, and precipitation, respectively. The amount of nitrogen per hectare and the precipitation in a season were divided into five levels respectively. Nitrogen and phosphorus leaching loss per soil type was classified as follows, enhanced grade, reduced grade and uncertain grade. Soil body configurations of nitrogen and phosphorus leaching loss were classified as enhanced and reduced grades. Acknowledgments. Project supported by the Science and Technology Projects of Qingdao (08-2-1-36-nsh and 09-1-1-53-nsh).
References 1. Xing, G.X., Shi, S.L.: Situation of nitrogen pollution in water bodies in SuZhou region. Acta Pedologica Sinica 38(4), 540–545 (2001) 2. Xiong, Z.Q., Xing, G.X.: Non-point N pollution of lakes, rivers and wells in the Taihu Lake region. Rural Eco-Enivironment 18(2), 29–33 (2002) 3. Xu, Q.X., Meng, Z.F., Yu, C.H.: Approaches for reduction of nitrate contamina¬tion on vegetable by appropriately applying fertilizers. Agro-environmental Protection 19(2), 109–111 (2000) 4. Corwin, D.I., Wagenet, R.J.: Application of the Modeling of non-point Sources Pollutants in the vadose zone. Journal Environment Quality 25, 403–411 (1996) 5. Griffin, J.R.: Introducing NPS Water Pollution. EPA Journal, 6–9 (November /December 1991) 6. Kronvang, B.: Diffuse nutrient losses in Denmark. Water Science Technology 133, 81–88 (1996) 7. Boersp, C.M.: Nutrient emissions from agriculture in the Netherlands: causes and remedies. Water Science Technology 33(1), 183–190 (1996)
Study on Vegetable Field Evaluation Index System for NPS Pollution
715
8. Feng, H.X., Yang, X., Li, X.Y., Wang, M.Z.: The effects of continuous cropping of vegetables on the biochemical properties of soil. Journal of ChangJiang Vegetables 11, 40–43 (2004) 9. Zhang, J.H., Wang, J.Q., Wan, Y., Han, C.: Agricultural non-point source pollution investigation and assessment in Qingdao. Chinese Agricultural Science Bulletin 26(10), 276–280 (2010) 10. Zhang, J.H., Wang, J.Q., Liu, J.T., Yao, Z.X., Li, D.P.: Evaluation of surface water pollution in Qingdao vegetable area based on gray correlation method. Shandong Agricultural Sciences 5, 78–82 (2010) 11. Zhou, G.Z., Zhang, J.H., Wang, J.Q., Li, J.: Application of the fuzzy mathematics in evaluation Dagu River water quality. Journal of Agro-environment Science 29(suppl.), 191–195 (2010) 12. Zhang, G.L., Zhang, S.: Advances of cropland Nitrogen Leaching. Soil 6, 291–297 (1998) 13. Liu, J.L., Li, R.G., Liao, W.H.: The yield response of vegetable to phosphate fertilizer and soil Phosphorus accumulation in a Chinese Cabbage-capsicum Rotation. Scientia Agricultura Sinica 38(8), 1616–1620 (2005) 14. Shi, C.Y., Zhang, F.D., Zhang, J.Q.: Change of soil nutrients under greenhouses under long-term fertilization condition. Plant Nutrition and Fertilizer Science 9(4), 437–441 (2003) 15. Johnes, P.J.: Evaluation and management of the impact of land use change on the nitrogen and phosphorus load delivered to surface waters, the export coefficient modeling ap¬proach. Journal of Hydrology 183, 323–349 (1996) 16. Chen, S.G.: Dry calcareous soil characteristics and nitrogen volatilization loss of ammonia Channels. Agricultural Research in the Arid Areas 3, 28–37 (1988) 17. Chen, X., Jiang, S.Q., Zhang, K.Z., Bian, Z.P.: Law of phosphorus loss and its affecting factors in red soil slopeland. Journal of Soil Erosion and Soil and Water 5(3), 38–41, 63 (1999)
Study on Water Resources Optimal Allocation of Irrigation District and Irrigation Decision Support System Liang Zhang1,2, Daoxi Li2, and Xiaoyu An2 1
2
Zhengzhou University, Zhenzhou, P. R. China North China University of Water Resources and Electric Power, Zhenzhou, P. R. China
[email protected] Abstract. This paper develops the system of optimal allocation of water resources in irrigation district and irrigation decision-making support which integrates technologies of decision-making support, information management, information search and so on. It has the multi-function of water production function calculation, crop water requirement calculation, water resources optimal allocation and real-time amending the decision of irrigation. This system integrates the experience of experts with computer technology to guide farmers to irrigate in a proper way, for which the limited water resources can produce a marked effect on irrigation, so irrigation district management and efficiency are improved. Keywords: Optimal allocation of water resources, irrigation district, watersaving irrigation, decision support system.
1 Introduction As we all know, agriculture is the main consumer of water. At present, agriculture of China faces water shortage, for agricultural water has been diverted by industrial water and domestic water, meanwhile, serious waste of agricultural water, lack of corollary irrigation facilities, sever defacement of trenches and long-term flooding irrigation lower the effective availability of irrigation water, as a result, valuable water can’t work in due course, therefore, we must vigorously advocate water-saving irrigation to construct a water-saving agriculture[1]. The key of water saving in agricultural irrigation is management[2], so management is meaningful in this area. There have been many studies about water-saving irrigation in China, but most of them are about hardware of facilities of water-saving irrigation, few about software such as water resources and optimal allocation[3]. To apply computer technology to irrigation district management will make full use of the valuable experience of experts in water conservancy with the help of modern technology to make a significant contribution to the improvement of irrigation district management, for which study on optimal allocation of water resources in irrigation district and irrigation decision supporting system is significant1. 1
Endowed by national science and technology supporting project (2007BAD88B02) (2006BAD11B09-2).
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part II, IFIP AICT 345, pp. 716–725, 2011. © IFIP International Federation for Information Processing 2011
Study on Water Resources Optimal Allocation of Irrigation District
717
2 A Brief Review of DSS Applied in Water Resource Management Decision support system (DSS) is a form of information system development, which is firstly pointed out by Scott Norton in 1971. It is an information system based on computer to support decision-making action, and a kind of interactive software aiming at assistant decision-making of application problems such as planning, management, scheduling, combat command and scheme optimization. With the developing of computer technology and the introducing of expert system and artificial intelligent technology, DSS technology is combined with other various technologies and has been developed into expert system and intelligent decision support system. DSS has been applied to water resource management since 1980s. Raboh.R.Reiter, J.and Gaschnig. J(1982) [4] presented HYDRO system to supply a similar parameter to watershed feature parameter which is selected by hydrological experts with great efforts and be applied to estimating the effect of different hydrological factors. Palmer. R. N and Tull, R. M(1987) [5], Palmer, P.N and Holmesk. J(1988) [6] have successively invented SID and WMS related to expert system of drought management plan, these two systems are similar in function and can be used for predicting and displaying the information related to drought management plan. Based on discrimination of the similar degrees of present drought and past drought, according to their experience, users can make decision of water quantity optimal distribution with linear programming model. Based on drought degree, Raman.H[7] has built up expert system with linear programming model to conduct crop optimization to guide drought scheduling decision that irrigation system will face in the future. CADSM[8] model is an expert system functioning as decision support, which can simulate crop yield and crop water-needing process to predict the effect of soil salinity and moisture on yield to supply users with water distribution plan of different canal systems. The software of water resource management decision support system in China is developed and applied relatively late. Wenbing Weng et al. (1992)[9] develope decision support system of water resource planning of JIng-Jin-Tang area (Beijing, Tianjin and Tanggu), this system has the function of expert knowledge and consultation. Jianxin Xu(1999)[3] developes regional water resource planning and irrigation water-saving irrigation development of Expert system through analyzing main factors of irrigation technology choice in irrigation area and introducing semi-structured and multi-objected optimum technique. Based on Penman’s formula, Zhouping Shangguan(2001)[10] combines with present agronomy knowledge of northwest arid area, model and experience to conduct system integration to build up intelligent decision support system by using artificial intelligent technology. Hujun Shang(2002)[11] points out developing model combined with data-based system , expert system and computer simulation through researching water-saving irrigation prediction and decision management database system. However, the studies mentioned above are mainly theoretical, exploratory or expert consulting; most results of them haven’t been perfect and practical yet. So far the study on water resource management and agricultural irrigation decision support system in China still exists in the exploring stage.
718
L. Zhang, D. Li, and X. An
3 Introduction of the System The developing and operating of water resources optimal allocation of irrigation district and irrigation decision support system in this study are based on Chinese Windows platform, main framework of this system software is compiled with Borland Delphi 7.0, using Delphi, some numerical computation modules are developed by data interface with the method of Matlab, the program of assistant decision database of water-saving irrigation is designed by Paradox 7.0 formula. This system is a system which has a database, a data management system and model calculation program library. It has a brief human-computer interface and a capability of reasoning and outputting the result of words and charts. Fig.1 indicates the system structure.
Fig. 1. System structure
The general function of the management system of database is to memorize, search, collate, collect, and survey all sorts of data. Meanwhile, this system can also supply necessary data for related results; this system can be divided into three parts in general, as follow: Database of fundamental data, which deposits all the basic data of irrigation districts including the proportion of them, crop planting, social economy condition of them, population information of them and so on. Text database, which deposits the data in form of text, mainly including calculation result memorized in form of text which calculated by working model. Chart database, which mainly deposits all sorts of charts of system, and can output the calculation result in form of charts according to the requests of the customers, by the way, the output information will be made more intuitionistic.
①
③
②
Study on Water Resources Optimal Allocation of Irrigation District
719
①
The reasoning system of model calculation contains three parts: Crop water production function calculation; Water resources optimal allocation of irrigation districts; design of optimal irrigation system under the condition of insufficient irrigation. These subparts not only can work independently but also can work together as a whole.
②
③
4 Key Technology of Model 4.1 The Constitution of Water Production Function and Parameters Calculation 4.1.1 Analysis of Water Production Function In order to conduct water resources optimal allocation to achieve the highest efficiency, we must set up the function of irrigation amount and yield of all sorts of crops in different hydrological years and make use of indirect relationship between irrigation amount and yield to get the function of irrigation amount and net benefit. In real course of solving, the function formula of irrigation amount—crop yield in different hydrological years is constituted firstly as usual, and then irrigation amount—irrigation benefit function as follow is constituted by using the relationship between input and output. By derivation of it, the change rate of benefit of crop in every unit of irrigation amount can be achieved, that is marginal benefit. According to marginal benefit, efficient irrigation method of limited water resources for crop can be inferred. The formula of water production function usually has the linear relationship, quadratic parabola relationship, power exponential function and so on. Based on a great amount of research result, and analysis of the experiment data of past irrigation experiments, the author thinks that as for the plain areas of the Yellow River, the Huaihe River and the Haihe River in China, water production function had usually better appear to be a quadratic parabola. 4.1.2 The Solving Principle of Water Production Function Supposing the basic formula as follow:
conclusion: B 2.6 Connection of Man-Machine IE6 is basic interface in this system, can make interaction with user by draw menus, toolbars, icons, graphics and tables, selective prompting be given, you can make selection by menus, the whole operation can be completed only by computer mouse and keyboard, then user can get the model and output the results. Mean while, the system provides the help files to make explanation for using the system. In addition, the contact surface of system has the good fault tolerance too, gives the error message and the processing prompt through examination common mistake by setting error trap, to ensure the correctness by user input.
3 Results and Discussion The system of cotton information management and cotton fertilization recommendation decision support system based on WEB realizes the main function such as :data management, system management, information query, fertilizer recommendation, soil evaluation, the expert knowledge and consultation, the result output and the system maintenance management and so on (Fig.2-5). 3.1 Data Management Module There are 5 attribute database management module in the module[5] (soil basic information database, fertilizer information database, fertilizer amount of previous years, fertilizer parameter database, user information database), each administration module has the operation functions such as: increase, deletion, saves, printing, search, sorting, screening and so on. and give the right of remote input and revise to user, advanced user may renew the attribute data whenever, so make the database can always reflect the newest tendency of farmland nutrients and other management information. Also user may carry on maintenance to database and module base(mainly to data edition, update and so on).At first, data be inputed in the database, so as to be called for inquiry and recommendation. 3.2 Information Inquiry Module Provides two ways to inquire for fertilizing scheme and field information. The user can copy the data from database according to the need, simultaneously the data in database can be batch introduced into through this contact surface. Information query and screening realizes information acquisiton function according to the query condition and screening condition the user combined to the data in database[5].The table for Inquiring or screening may be the table of expert system standard, also may be the result which the system recommendation decision-making leaves, the data message inquire may be printed directly and derive.(Fig.2)
730
Y.-m. Dang and X. Lv
Fig. 2. Information Query
Fig. 3. Management of Soil evaluation
Study on Web-Based Cotton Fertilization Recommendation
731
3.3 Module of Soil Appraisal Primary to compute membership degree of Organic Matter, available nitrogen, available P, available N, and obtains the index value of comprehensive evaluation. (Fig.3) 3.4 The Design of Soil Fertilizer Recommendation Module The module including conventional fertilization recommendation by soil test, drip irrigation fertilizer recommendation, effect function recommendation, organic fertilizer recommendation and microelements fertilizer recommendation. Under the suggestion of guide(Fig.4),the user can input the data of soil nutrient, fill in fertilizer amount of strip field, variety selection of fertilization and establish goal output according to local actual situation. The system judges whether the data user input is reasonable based on the ordinary years’ data of meteorology, soil, variety and so through operating the knowledge model, if reasonable, then make fertilizer recommended by calling the knowledge module of fertilizer module, according to the information user filled in, the result of fertilizer recommendation can be printed on the formula to apply fertilizer card or data export by Excel, the user can modify, edits and prints the recommendation result by Excel datas exported. If unreasonable, the system modify the plan which have be made, and sends it to the fertilizer model to make forecast, so circulates, until generating a set of fertilizer recommendation plan which meet requirement[6].
Fig. 4. Frame work of fertilize recommendation
732
Y.-m. Dang and X. Lv
3.5 Maintenance and Management of System The system gives different jurisdiction to the different grades user, the user may browse, inquire, modify, increase and delete the knowledge of the knowledge base and datas of database in own purview.( Fig.5)
Fig. 5. Management and maintenance of database
4 Conclusion Precise fertilization is important part of Precision Agriculture, It is the best fertilizer plan established above the scientific method to fertilizer.This system construct comprehensive digital and intelligent decision support system based on WEB using the SQL+JSP+Win2000, and collected and arrangemented soil information of all regiments by Internet, and provides the accurate and reliable soil material, then obtained the comprehensive index of soil fertility using fuzzy mathematics principle, make fertilizer recommendation, establish fertilization model of soil nutrient, the fertility district and the formula district of crop specific fertilizer; establish the balance fertilization system of nitrogen, the phosphorus, the potassium and the trace element according to the consideration about soil supplying nutrient capability
Study on Web-Based Cotton Fertilization Recommendation
733
and fertilizer needs of crops. It will be important theory value and the practice significance to implementing balance fertilization technique and generalization of other Scientific Research Achievements.
References [1] Zhu, Y., Cao, W.x.: A Knowledge Model-and Growth Model-Based Decision Support System for Wheat Management. Sceintia Agricultura Sinica 37(6), 814–820 (2004) [2] Hao, Y.-l., et al.: Programming Technology of J2EE. The Publishing House of Qinghua university and Beijing Jiaotong University, Beijing (2005) [3] Xiong, F.l., Qiao, K.z., Hu, H.-y.: Agricultural Expert System and Development Tools. The Publishing House of Qinghua University, Beijing (1999) [4] Yang, B.-r.: Knowledge Engineering and Knowledge Discovery, pp. 423–450. The Publishing House of Metallurgy Industry, Beijng (2000) [5] Yan, C., Lv, X.: Information Management and Fertilizing Decision System Based on Soil Nutrient of SuperMap IS Cotton Field. XinJiang Agricultural Sciences 41(6), 427–430 (2004) [6] Xie, K.-w., Cheng, Y.-q.: The Design and Implementation of the Expert System Based on Database. Journal of Hubei Institute For Nationalities (Natural science) (2), 193–196 (2005)
Author Index
Ai, Jumei III-721 An, Xiaoyu II-716 An, Zhengguang I-729, I-737 Bai, Jun-Hua II-90, IV-16 Bai, Wei III-92 Bai, Yichuan III-572 Bai, Zhongke III-173 Bao, Jie III-725, IV-610 Bao, Nisha III-173 Bao, Wenxing I-48, II-124, III-132, III-464, III-491 Bi, Chunguang I-312 Bi, Lan IV-450 Bonifazi, Giuseppe IV-751 Cai, Guoyin II-243 Cai, Hongzhen III-341 Cai, Kewen IV-390 Cai, Lecai I-539 Cai, Tijiu II-682 Cao, Hongxin II-441 Cao, Qinghua IV-237 Cao, Qing-Song IV-410, IV-450 Cao, Ran II-322 Cao, Shehua IV-192 Cao, Weixin III-186 Cao, Weixing I-446, II-479 Cao, Wenqin III-14 Cao, YongSheng II-110 Chang, Ruokui III-106 Chang, Zhongle I-590 Chao, Liu II-425 Che, Zhenhua I-569, II-198 Chen, Aixi IV-89 Chen, Baisong II-525 Chen, Baorui I-250, II-658, IV-134 Chen, Bin III-123 Chen, Bing IV-602 Chen, Di-yi II-205 Chen, Guifen I-312, II-408 Chen, Hong II-61 Chen, Hongjiang IV-215 Chen, Hui IV-268 Chen, Jianhua I-569, II-198
Chen, Jin II-607 Chen, Kelou II-139 Chen, Lairong III-532 Chen, Lidong I-359 Chen, Liming II-505 Chen, Ling I-178 Chen, Liping I-103 Chen, Tian’en I-103 Chen, Yaxiong I-390, III-357 Chen, Yongxing III-92 Chen, Zhaoxia IV-63 Cheng, Jianqun IV-361 Cheng, Jihong III-222 Cheng, Jilin II-283, III-554 Cheng, Youping I-149, I-353 Chu, Changbao IV-237 Ci, Xin IV-345 Cohen, Oded I-630 Cui, Hongguang I-267, I-428 Cui, Weiwei II-674 Cui, Yunpeng I-56, II-573, III-648 Dai, Lili I-267, I-428 Dai, Rong II-148 Dan, Nie III-732 Dang, Yv-mei II-726 Deng, Guang I-304 Deng, Hubin IV-96, IV-376 Deng, Jinfeng II-211 Deng, Lei IV-255 Deng, Li II-339, III-390 Deng, Shangmin I-674 Diao, Haiting III-57 Ding, Chao II-11 Ding, Jianjun III-36 Ding, Li I-437 Ding, Qingfeng IV-279 Ding, Qisheng IV-629, IV-642, IV-650, IV-659 Ding, Wenqin I-456 Ding, XiaoLing IV-345 Dong, Jing I-35, II-561 Dong, Jingui I-576 Dong, Lihong IV-231
736
Author Index
Dong, Qizheng IV-167 Dong, Shiyun IV-231 Dong, Sufen II-365 Dong, Yiwei III-92 Dou, Yantao IV-78 Du, Bin IV-355 Du, Huibin I-487 Du, Jing IV-720 Du, Jun I-155 Du, Mingyi I-681, II-243 Du, Shuyuan I-16 Duan, Qingling IV-691 Duan, Qingwei IV-134 E, Yue
II-110
Fan, Honggang I-35 Fan, Shijuan IV-116 Fan, Xinzhong I-576, I-590 Fang, Hui IV-124 Fang, Junlong IV-616 Feng, Shaoyuan II-473 Feng, Yaoze IV-184 Fu, Bing III-186 Fu, Qiang III-419 Fu, Xueliang I-487, I-526 Fu, Yu II-131 Fu, Zetian IV-672, IV-680 Fu, Zhuo II-525 Gan, Weihua II-400, II-579 Gao, Bingbo II-415 Gao, Gaili III-604 Gao, Haisheng IV-355 Gao, Hongyan II-53 Gao, Lingwang I-594 Gao, Miao I-282 Gao, Rui I-138 Gao, Shi-Ju IV-16 Gao, Xiaoliang III-732 Gao, Yang III-539 Gao, Ying IV-260 Gao, Yun I-600, II-61 Gao, Zhi-Fan IV-410 Ge, Daokuo II-441 Ge, Ningning I-594 Geng, Duanyang II-158 Geng, Duayang II-531 Geng, Xia III-1
Geng, Zhi III-710 Gitelson, Anatoly IV-47 Gong, Bikai IV-602 Gong, Shuipeng IV-616 Gong, Yi III-554 Gu, Jingqiu III-661 Gu, Wenjuan IV-543 Gu, Xiaohe I-296 Gui, Dongwei I-321 Guo, Huiling II-18, II-491 Guo, Mingming I-409, III-327 Guo, Qian I-374, III-222 Guo, Rui II-551, III-357 Guo, Wei II-322 Guo, Yuming I-691 Han, Jinyu III-539 Han, Ping I-282, I-290 Han, Qiang I-48 Han, Xiangbo I-16, I-472, I-717 Hannaway, David B. II-441 He, Bailin IV-63 He, Bei I-138 He, Bin II-18 He, Dongxian III-725, IV-504, IV-610 He, Fen II-102 He, Feng IV-8 He, Jianbin I-35, II-561 He, Junliang I-519 He, Lian II-283 He, Pengju II-573, III-648 He, Qingbo IV-206 He, Renwang II-517 He, Tian III-316, III-375 He, Wenying II-551, III-357 He, Yong IV-124 Hu, Chunxia I-238, II-41 Hu, Haiyan I-41, III-158 Hu, Jianping I-401, I-456, I-555, II-496, III-249 Hu, JinYou I-623, III-656 Hu, Juanxiu IV-504 Hu, Kaiqun III-304, III-483 Hu, Lin III-138 Hu, Ping II-30 Hu, Runwen II-667 Hu, Siquan I-131, IV-71 Hua, Yu II-650 Huang, Caojun II-309 Huang, Chong I-582
Author Index Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang, Huang,
Guanhua I-643, II-185 Han III-572 Kelin IV-89 Lan III-289 Qing I-250, II-658 Sheng II-351 Wen III-598 Wenjiang I-296, III-280 Xiaomao I-25 Yan II-682 Yanguo IV-321 Ying I-210 Yingsa I-401, I-456, II-496 Yinsa III-249 Yuxiang II-351 Zhigang IV-306
Ji, Baoping III-84 Ji, Ronghua III-304, III-483, III-532 Ji, Ying I-138 Jia, Chaojie I-390, III-357 Jia, Guifeng IV-198 Jia, Shaorong III-198 Jia, Song II-700, III-41 Jiang, Haiyan II-479, III-186 Jiang, Huanyu I-729, I-737 Jiang, Lihua I-149, I-353 Jiang, Na II-473 Jiang, Qiuxiang III-419 Jiang, Wencong II-381 Jiang, Xi II-706 Jiang, Xiangang IV-30 Jie, Dengfei II-118 Jin, Dan III-347, III-445 Jin, Tingxiang I-238, II-41 Jinbin, Li I-335 Jinfu, Lu IV-563 Jinlong, Lin I-608 Juanxiu, Hu III-725 Kai, Wang II-425 Kan, Daohong IV-701 Kong, Fanrang IV-206 Kong, Wenwen IV-124 Kuang, Tangqing IV-543 Lai, Zhigang I-508, IV-361 Lei, Jiaqiang I-321 Lei, Wen I-539 Lei, Xiaojun II-479
Li, Li, Li, Li, Li, Li, Li, Li,
737
Baojun II-30 Biao I-590 Changyou I-487, I-526 Chen III-549 Chengyun I-227, I-335 Chunzhi IV-147 Cunjun I-296, III-280 Daoliang III-725, IV-610, IV-629, IV-642, IV-650, IV-659, IV-672, IV-680, IV-701, IV-710, IV-720, IV-727, IV-735, IV-742 Li, Daoxi II-716 Li, Dapeng II-706 Li, Deying IV-474, IV-514 Li, Fanghua II-682 Li, Fengmin II-551 Li, Gailian I-238, II-41 Li, Gang I-250, II-658, IV-134 Li, Guo IV-691 Li, Guoqing III-241 Li, Haifeng I-321 Li, Hengbin III-379 Li, Honghui I-526 Li, Hongjian IV-89 Li, Hongwen IV-720 Li, Hongyi III-212 Li, Hui I-594, II-317, III-304, III-483 Li, Jia wei IV-39 Li, Jianyun IV-474 Li, Jin III-580 Li, Jing II-90, IV-467, IV-528 Li, Jun IV-382, IV-521 Li, Lin I-35, II-309, II-561 Li, Ling I-698 Li, Linyi II-587 Li, Lixin I-508 Li, Manman III-629 Li, Maogang III-20, III-29 Li, Meian I-487 Li, Minghui IV-514 Li, Mingyong I-576 Li, Na IV-480, IV-537 Li, Peiwu I-600, IV-246 Li, Qiaozhen III-92 Li, Qingji III-413, III-440 Li, Qingqing I-16 Li, Shao-Kun IV-16 Li, Shaokun II-90, II-691 Li, Shijuan I-219, I-261, I-476 Li, Shuqin II-351
738
Author Index
Li, Wei II-102 Li, Wenxin III-572 Li, Wenyue III-598 Li, Xianyue I-155 Li, Xiaoqin IV-294 Li, Xiaoyu I-600, II-61, IV-184, IV-246 Li, Xinlei I-409, III-327 Li, Xuemei II-18 Li, Yan II-185 Li, Yanling II-381, II-392 Li, Yaoming II-607 Li, Yi I-594 Li, Ying III-57 Li, Yong IV-221 Li, Yuan III-500, III-539 Li, Yuanzhang I-68 Li, Yuhong I-275, I-711, IV-575 Li, Yunkai I-155 Li, Yuzhong III-92 Li, Zengyuan I-304 Li, Zhigang III-500, III-539 Li, Zhihong II-465, III-563, III-572 Li, Zhimei III-563, III-572 Li, Zhizhong III-704 Li, Zhongqi IV-177 Liang, Jing II-633 Liang, Qing IV-30 Liang, Yong I-547, II-381, II-392, II-700, III-1, III-41, III-390, III-403, III-452 Liang, Yusheng III-57 Liao, Weichuan II-259 Liao, Xinglong I-1, I-532 Liming, Lu IV-563 Lin, Fengtao IV-568 Linker, Raphael I-630 Liu, Baifen IV-260 Liu, Changju IV-184 Liu, Chengliang II-23 Liu, Cuie III-8 Liu, Cuiling II-317 Liu, Ergen IV-1, IV-390 Liu, Fa I-401, I-456, II-496 Liu, Fei IV-124 Liu, Gang I-138, I-409, III-327, III-580 Liu, Guiyuan IV-96 Liu, Haijun II-185 Liu, Hailong II-290 Liu, Hua III-106
Liu, Jianshu IV-494 Liu, Jianting II-706 Liu, Jie I-600, IV-246 Liu, Jingbo II-415 Liu, Jingyu I-68 Liu, Jingyuan II-465 Liu, Jiping II-674 Liu, Juanjuan IV-108 Liu, Jun III-604 Liu, Junming III-629 Liu, Leping IV-167, IV-333 Liu, Li-Bo I-62 Liu, Lin I-227, I-335 Liu, Liyong IV-727, IV-735, IV-742 Liu, Lu I-623 Liu, Min IV-345 Liu, Mingzeng II-30 Liu, Muhua II-434, IV-467, IV-528 Liu, Ping’ an I-508, IV-361 Liu, Pingan IV-306 Liu, Shengping I-476 Liu, Shihong I-56, II-573, III-158, III-179, III-648 Liu, Shuangxi III-379, III-620, IV-710 Liu, Tao IV-427 Liu, Wei I-178, II-18 Liu, Xiaodong IV-376 Liu, Xiaojun I-446 Liu, Xiuping II-30 Liu, Xu III-289 Liu, Xue IV-672 Liu, Xuming III-704 Liu, Yajuan IV-103 Liu, Yan II-441 Liu, Yande II-1, III-613, IV-427 Liu, Yang I-681, II-243 Liu, Yanqi II-30 Liu, Yin III-327 Liu, Ying II-706 Liu, Yongbin IV-206 Liu, Yongxia II-441 Liu, Yu-xiao II-205 Liu, Zhanli I-472, I-717 Liu, Zheng II-177 Liu, Zhengfang IV-89 Liu, Zhengping IV-108 Liu, Zhifang I-87 Liu, Zhimin IV-89 Liu, Zhipeng I-721
Author Index Liu, Zhongqiang I-76, III-46, III-682, III-696 Long, Changjiang I-25, I-195 Long, Yan II-205 Lu, Anxiang I-282, I-563, II-83 Lu, Daoli III-123 Lu, Gang II-11, III-8 Lu, Huishan I-729, I-737 Lu, Jiahua I-569, II-198 Lu, Peng II-11, II-329, III-8 Lu, Quanguo IV-237 Lu, Shaokun III-725, IV-610 Lu, Weiping I-275, I-711, IV-575 Lu, Yan-Li IV-16 Lu, Zhixiong IV-294 Lu, Zhongmin II-357 Luan, RuPeng II-615 Luan, Xin I-590 Luan, Yunxia II-83, II-457 Luo, Chagen IV-368 Luo, Changshou III-638, III-672 Luo, Chunsheng IV-467, IV-528 Luo, Laipeng IV-1 Luo, Qingyao III-710 Luo, Shimin IV-286 Luo, Xiaoling IV-255 Lv, Jiake III-512 Lv, Xin II-290, II-726 Lv, Yongliang II-706 Ma, Daokun IV-629, IV-650, IV-659 Ma, Hailei I-526 Ma, Juncheng IV-680 Ma, Li II-408 Ma, Liang II-538 Ma, Lili IV-616 Ma, Liuyi II-597 Ma, Lizhen III-106 Ma, Xiaoguang II-465 Ma, Xiao-yi II-205 Ma, Xinming I-437, I-614, II-357, III-269 Ma, Xu I-1, I-532 Ma, Yuan IV-333 Ma, Zhihong I-282, II-83, II-234, III-592 Mao, Enrong III-257 Mao, Hanping II-53 Mao, Shuhua III-721 Mei, Weng II-650
739
Men, Weili I-674 Meng, Haili III-598 Meng, Hong III-158 Meng, Qingyi II-473 Meng, Xianxue III-179 Miao, Pengbo IV-528 Min, Shungeng III-592 Ming, Bo II-691 Mingyin, Yao I-608 Muhua, Liu I-608 Naor, Amos I-630 Ning, Dongzhou IV-376 Ouyang, Aiguo
IV-368
Pan, Fangting III-231 Pan, Guiying II-71 Pan, Jiayi III-710 Pan, Juan I-367 Pan, Ligang I-282, I-290, I-563, II-83, II-234, II-457 Pan, Qilong IV-735 Pan, Yuchun II-525 Pang, Siqin IV-78 Pei, Chunmei II-18, II-491 Peng, Bo I-119 Peng, Cheng II-641, III-661 Peng, Lin I-417 Ping, Hua I-290, II-234, II-457 Ping, Jia III-428 Ping, Xuecheng IV-306 Qi, Kun II-61 Qi, Lijun III-304, III-483 Qi, Limeng II-339 Qi, Long I-1 Qiao, Hongbo II-650 Qiao, Xiaojun III-66, III-75 Qiao, Xibo I-576, I-590 Qiao, Zhong III-473 Qin, JiangLin IV-47 Qin, Xiangyang I-563, III-580 Qing, Chang I-335 Qing, Zhaoshen III-84 Qiu, Wanying II-521 Qiu, Xiaobing III-473 Qiu, Ying IV-420 Qiu, Yun II-300, III-113, III-138 Qiulian, Li I-608
740
Author Index
Rao, Guisheng II-339 Rao, Honghui III-613 Ren, Jiwen IV-494 Ren, Rong III-132 Ren, Shumei I-155 Ren, Souhua II-309 Ren, Wentao I-267, I-428 Ren, Yanna II-357, III-269 Rundquist, Donald IV-47 Serranti, Silvia IV-751 Shang, Huaping II-227 Shao, Xiuping I-401, I-456, II-496 Shao-Wen, Li II-425 She, Chundong I-131, IV-71 Shen, Changjun IV-435 Shen, Lifeng IV-720 Shen, Tao IV-30, IV-592 Shen, Zuorui I-594 Shi, Guoqing III-231 Shi, Liang II-124 Shi, Xiaoxia II-264, II-641 Shi, Yan III-20, III-29 Shi, Yinxue III-367 Shi, Yuanyuan I-614 Shi, Yuling III-289 Shi, Zhou II-71 Shu, Xiaoping IV-521 Si, Yongsheng I-138 Song, Mingyu III-106 Song, Qin I-125 Song, Xiaoqiang III-322 Song, Xiaoyu I-296 Song, Yunliang III-123 Song, Zhenghe III-257 Steele, Mark IV-47 Su, Xiaolu I-41 Su, Yuan I-227, I-335 Sui, Xueyan II-691 Sun, Chao III-36 Sun, Chengli IV-8 Sun, Fa-Xiong IV-450 Sun, Guojun I-390, II-551, III-357 Sun, Jiang I-563 Sun, Jinping III-403, III-452 Sun, Jinying II-441 Sun, Kaimeng II-218 Sun, Li III-198 Sun, Ming IV-39 Sun, Nan II-465
Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun, Sun,
Ruizhi III-367 Sufen I-56, III-638, III-672 Suping I-576 Wenbin III-57 Wensheng III-165 Xia I-16 Xiang III-661, IV-583 Xiaoqing II-691 Xudong II-1 Yonghua I-96, I-464 Yongxiang I-547, III-1 Zhiguo I-9 Zhongwei II-329, II-374
Tai, Haijiang IV-642, IV-650 Tan, Feng II-479 Tan, Jinghe I-590 Tan, Jingying III-347, III-445 Tan, Yu-an I-68 Tan, Zongkun IV-47 Tang, Bin IV-89 Tang, Chengwen IV-198 Tang, Liang II-479 Teng, Guanghui III-704 Teng, Guifa II-365 Teng, Yun II-682 Wan, ChangZhao III-222 Wan, Peng I-25, I-195 Wan, Shubo III-146 Wan, ShuJing III-403 Wang, Bai II-682 Wang, Baoqing II-441 Wang, Buyu I-487, I-526 Wang, Changsheng II-30 Wang, Cheng III-66, III-84 Wang, Chun II-345, II-567 Wang, Dong III-592 Wang, Dongqing I-487 Wang, Fang IV-691 Wang, Fang-Yong IV-16 Wang, Fangzhou III-165 Wang, Fei III-198 Wang, Fengxin II-185 Wang, Fuxiang III-563 Wang, Haiguang I-582 Wang, Haiou I-131, III-231, IV-71 Wang, Hongbin I-35, II-561 Wang, Jian I-62, II-415, III-113 Wang, Jianqin I-119
Author Index Wang, Jihua I-282, I-290, I-563, II-83, II-525 Wang, Jing III-249 Wang, Jinhua IV-376 Wang, Jinxing III-379, III-620, IV-710 Wang, Junfeng I-131, IV-71 Wang, Junqiang II-706, IV-592 Wang, Kaili I-643 Wang, Kaiyi I-76, III-46, III-682, III-696 Wang, Ke-Ru IV-16 Wang, Lianzhi IV-629 Wang, Li-jun III-99 Wang, Ling II-290 Wang, Lingyan I-155 Wang, Pu III-491 Wang, Qiang III-269 Wang, Qing III-347, III-445 Wang, Qingchun III-532 Wang, Qiong IV-16 Wang, Ruijuan I-374 Wang, Sangen II-166 Wang, Shengfeng III-428 Wang, Shicong III-473 Wang, Shufeng I-76, III-696 Wang, Shushan III-123 Wang, Shuwen IV-616 Wang, Shuyan I-110 Wang, Susheng I-359 Wang, Wei I-96, I-464, I-600, IV-78, IV-184, IV-246 Wang, Wensheng I-9, I-149, I-203, I-353 Wang, Xi II-567 Wang, Xiangyou I-16, I-472, II-158, II-531 Wang, Xiao IV-467, IV-528 Wang, Xiaojun II-392 Wang, Xiaoli II-623 Wang, Xihua III-36 Wang, Xin I-381 Wang, Xingxing III-20 Wang, Xinzhong II-567 Wang, Xu IV-134 Wang, Xuan III-512 Wang, Yanan III-92 Wang, Yang II-329, II-374 Wang, Yangqiu II-491 Wang, Yanlin IV-514 Wang, Yuanhong III-106 Wang, Yunsheng I-374, II-434, III-222
Wang, Zhaopeng I-576 Wang, Zhenzhi III-46, III-682 Wang, Zhiwei IV-116 Wang, Zhongyi III-289 Wang, Zilong III-419 Wei, Chaofu III-512 Wei, Enzhu III-249 Wei, Lin III-563 Wei, Qingfeng III-638, III-672 Wei, Xinhua II-607 Wei, Xiufang II-441 Wei, Yaoguang IV-642, IV-650 Wei, Yong III-106 Wen, Nannan IV-659 Wu, Chaohui I-275, I-711, IV-575 Wu, Dake II-166 Wu, Dan IV-390 Wu, Ding-Feng I-62, III-113 Wu, Dongsheng IV-382, IV-521 Wu, Hongchao I-576, I-590 Wu, Honggan I-304 Wu, Huarui III-661, IV-583 Wu, Jiajiao III-563 Wu, Jingzhu II-317 Wu, Qingping IV-89 Wu, Qiulan I-547, III-1 Wu, Quan III-198 Wu, Wenbiao III-75 Wu, Xiaoying II-465 Wu, Yali I-691 Wu, Yongchang II-264 Wu, Zhigang III-572 Xi, Junmei IV-237 Xi, Lei I-437, III-269 Xia, Hui I-9 Xia, Junfang II-667 Xia, Lianming II-158, II-531 Xia, Xiaobin IV-8 Xiang, Ling I-290, II-234 Xiang, Quanli I-267, I-428 Xiang, Xinjian I-495 Xiao, Chun-Hua IV-16 Xie, Dandan II-517 Xie, Deti III-512 Xie, Fengyun IV-443 Xie, Nengfu I-149, I-203, I-353 Xie, Rui-Zhi IV-16 Xie, Sanmao IV-314 Xie, Zuqing I-555
741
742
Author Index
Xin, Xiaoping I-250, II-658, IV-134 Xing, Qirong IV-177 Xing, Yajuan I-487 Xing, Zhen IV-435 Xiong, Bangshu IV-8 Xiong, Benhai III-710 Xiong, Guangyao IV-514 Xiong, Guo-Liang IV-410 Xiong, Jinhui I-110 Xiong, Shuping I-614, III-269 Xu, Beili IV-521 Xu, Binshi IV-231 Xu, Chunying III-92 Xu, Fuhou II-597 Xu, Hongmei I-656, I-698 Xu, Jian I-367 Xu, Jianxin III-428 Xu, Li III-473 Xu, Liming II-505 Xu, Ling I-227 Xu, Lunhui IV-321 Xu, Shenghang IV-108 Xu, Shipu II-434, III-222 Xu, Xianfeng IV-63 Xu, XiangBin IV-460, IV-486 Xu, Xiaoli IV-78 Xu, Xin I-437, II-357 Xu, Xingang I-296, III-280 Xu, Yizong I-210 Xue, Fengchang IV-623 Xue, Heru I-502, II-252 Xue, Long IV-403, IV-467, IV-528 Xue, Yan I-219 Xue, Yandong I-155 Yan, Congcong I-472 Yan, Hua III-75 Yan, Jianwu IV-237 Yan, Junyong IV-147 Yan, Lin II-633 Yan, Manfu I-87, I-343 Yan, Qin III-452 Yan, Xiaomei III-620 Yan, YinFa IV-345 Yan, Yuchun IV-134 Yane, Duan II-274 Yang, Chao IV-116 Yang, Deyong I-555, II-496 Yang, Fei III-327 Yang, Feng I-76, III-46, III-682, III-696
Yang, Fengping IV-177, IV-279 Yang, Guixia I-250, II-658, IV-134 Yang, Hao III-280 Yang, Huiying II-185 Yang, Jianhua II-131 Yang, Jing I-227, I-335 Yang, Juan I-374, III-222, III-269 Yang, Le I-569, II-198 Yang, Liang III-710 Yang, Linnan I-417 Yang, Min III-123 Yang, Minghao III-367 Yang, Peiling I-155 Yang, Ping III-413 Yang, Po IV-504 Yang, Sen II-551, III-357 Yang, Shuqin III-428 Yang, Tao III-298, III-322, III-413, III-440 Yang, Wenzhu IV-701, IV-710 Yang, Xiaodong III-280 Yang, Xiaohui II-351 Yang, Xiaorong I-149, I-203, I-353 Yang, Xiaoxia II-700, III-1, III-41 Yang, Xin I-275, I-711, IV-575 Yang, Xiuqing II-18, II-491 Yang, Yafei II-650 Yang, Yang III-316, III-375, IV-398 Yang, Yi I-267, I-428 Yang, Yibo II-517 Yang, Ying I-119 Yang, Yong I-35, I-110, II-561, III-598 Yang, Yongsheng III-464 Yang, Yujian III-146 Yang, Yushu II-322 Yao, Jie II-365 Yao, Shan I-367 Yao, Zhenxuan II-706 Ye, Baoying III-173 Ye, Fan IV-321 Ye, Hairong II-491 Ye, Shengfeng III-592 Yin, Jinju IV-494 Yin, Zhongdong II-538 Ying, Yibin I-729, I-737 Yong, Wang II-650 You-Hua, Zhang II-425 Yu, Feng II-615 Yu, Ligen III-704
Author Index Yuan, Haibo III-20, III-29 Yuan, Jun IV-294 Yuan, Junjing III-341 Yuan, Shengfa I-656, IV-198 Yuan, Tao II-434, II-587 Yuan, Xiaoqing IV-727, IV-742 Yuan, Xue III-304, III-483 Yuchuan, Yang IV-563 Yue, E. I-476 Zang, Yu III-257 Zang, Zhiyuan I-594 Zejian, Lei I-608 Zeng, Fanjiang I-321 Zeng, Qingtian I-203 Zeng, Yanwei II-381, II-392 Zeng, Yi III-241 Zeng, Zhixuan II-525 Zha, Xiaojing IV-382, IV-521 Zhang, Baihua III-525 Zhang, Baohui IV-134 Zhang, Baojun II-441 Zhang, Benhua I-428 Zhang, Changli IV-616 Zhang, Chengming III-390, III-403, III-452 Zhang, Chi I-103 Zhang, Chunlei II-441 Zhang, Chunmei III-29 Zhang, Chunqing III-620 Zhang, Dalei III-403 Zhang, Dongxing III-604 Zhang, Fan II-365 Zhang, Feng III-413 Zhang, Guoliang I-594 Zhang, Haihong I-721, II-118 Zhang, Hailiang II-1 Zhang, Hao I-437, II-357 Zhang, Haokun I-502 Zhang, Hong I-96, I-464, I-539 Zhang, Hongbin I-250, II-658, IV-134 Zhang, Jian I-623, III-341, III-656, IV-474, IV-537 Zhang, Jianhang I-343 Zhang, Jianhua III-304, III-483 Zhang, Jing I-381, I-698 Zhang, Jingjing I-623, III-656 Zhang, Jinheng II-706, IV-592 Zhang, Jishuai II-650 Zhang, Juan II-357
743
Zhang, Jun IV-246 Zhang, Junfeng I-56, II-615, III-638, III-672 Zhang, Junxiong II-102 Zhang, Kai II-23 Zhang, Lei IV-96 Zhang, Li II-357 Zhang, Liang II-716 Zhang, Lihua III-554 Zhang, Limin I-417 Zhang, Lingxian IV-672, IV-680 Zhang, Lingzi IV-691 Zhang, Longlong III-269 Zhang, Mei III-186 Zhang, Min I-569, II-198 Zhang, Mingfei IV-629 Zhang, Na I-367 Zhang, Ping IV-206 Zhang, Qing I-343 Zhang, Qingfeng II-158, II-531 Zhang, Rentian III-554 Zhang, Runqing II-339 Zhang, Shujuan I-721, II-118 Zhang, Shuyuan I-519 Zhang, Tingting II-400, II-579 Zhang, Tongda II-597 Zhang, Wei II-615 Zhang, Xiandi I-76, III-46, III-682, III-696 Zhang, Xiaodong II-53, II-691 Zhang, Xiaojing II-491 Zhang, Xiaolan IV-480, IV-537 Zhang, Xiaoyan III-146 Zhang, Xin III-66, III-75, IV-435, IV-701, IV-710 Zhang, Xu I-304 Zhang, Xuelan I-68 Zhang, Yang III-316, III-375 Zhang, Yanrong II-139 Zhang, Yaoli II-641 Zhang, Yu IV-221 Zhang, Yue I-227 Zhang, Yunhe III-66 Zhang, Yuou I-446 Zhang, Yuxiang II-597 Zhang, Zhen I-359 Zhao, Chunjiang I-76, III-580 Zhao, Dongjie III-289 Zhao, Dongmei III-473 Zhao, Fukuan I-125
744
Author Index
Zhao, Hu II-166 Zhao, Huamin I-721 Zhao, Huizhong I-569, II-198 Zhao, Jianshe III-598 Zhao, JiChun II-615 Zhao, Jingyin I-374, II-434, II-587 Zhao, Jingying III-222 Zhao, Lanying IV-294 Zhao, Liu I-290, II-83, II-234 Zhao, Longzhi IV-474, IV-480, IV-514, IV-537 Zhao, Mingjuan IV-480, IV-537 Zhao, Peng I-614 Zhao, Suolao II-441 Zhao, Ting II-252 Zhao, Wei I-155 Zhao, Wen IV-333 Zhao, Wenlong I-390 Zhao, Wenping I-16 Zhao, Xiaoming I-669 Zhao, Yanqing IV-398 Zhao, Yanru I-721 Zhao, Yujun II-561 Zhao, Yuling II-517 Zhao, Zhiyong II-177 Zheng, Guang I-437 Zheng, Huaiguo I-56 Zheng, Huoguo III-158, III-179, III-648 Zheng, Lihua III-473 Zheng, Meizhu IV-514 Zheng, Wengang III-66, III-75, IV-435 Zheng, Wenxiu III-379 Zheng, Yanxia I-519, II-177, II-473 Zheng, Youfei I-381 Zheng, Yuelan II-607 Zheng, Yujun I-125 Zheng, Zhihong II-23 Zhong, Guangrong III-231 Zhong, Mingdong IV-167 Zhong, Shiquan I-275, I-711, IV-575 Zhong, Zhiyou I-569, II-198 Zhou, Chao III-613, IV-368 Zhou, Dongsheng III-298 Zhou, Ermin II-139 Zhou, Fengqi IV-89
Zhou, Guo-Min I-62, II-300, II-623, III-113, III-138 Zhou, Huamao III-613 Zhou, Huilan IV-555 Zhou, Ji-Hui IV-443, IV-450 Zhou, Lianqing II-71 Zhou, Liying III-672 Zhou, Mingyao I-359 Zhou, Nan III-473 Zhou, Wei IV-246 Zhou, Weihong III-357 Zhou, XinJian IV-486 Zhou, Zexiang I-68 Zhou, Zhisheng II-309 Zhou, Zhu I-600, IV-246 Zhu, Chuanbao II-441 Zhu, Dawei II-441 Zhu, Dazhou I-563, III-84, III-92 Zhu, Dongnan I-526 Zhu, Fengmei IV-355 Zhu, Haiyan III-14 Zhu, Huaji III-661, IV-583 Zhu, Jianhua III-146 Zhu, Jie IV-410 Zhu, Juanuan II-441 Zhu, Qixin IV-279 Zhu, Shiping II-633 Zhu, Wei IV-486 Zhu, Wenquan I-681 Zhu, Yan I-446, II-479, III-186 Zhu, YePing II-110 Zhu, Yeping I-219, I-261, I-476 Zhu, Youyong I-227, I-335 Zhu, Yuwei II-400, II-579 Zhu, Zhenlin II-691 Zhu, Zhongkui IV-206 Zhu, Zhongxiang III-257 Zhuang, Weidong II-345, II-567 Zong, Li I-656, I-698 Zong, Wangyuan I-96, I-464 Zou, Qiang IV-124 Zude, Manuela III-84 Zuo, Changqing II-538 Zuo, Tingting II-131 Zuo, Yanjun I-1, I-532