Study on the optimal algorithm prediction of corn leaf component information based on hyperspectral imaging |
| |
| Institution: | 1. College of Information and Electrification Engineering, Shenyang Agricultural University, Shenyang 110866, PR China;2. Beijing Research Center for Agri-food Testing and Farmland Monitoring, Beijing 100097, PR China;3. Beijing Research Center of Intelligent Equipment for Agriculture, Beijing 100097, PR China;1. College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China;2. Engineering Research Center of Intelligent PSE, Ministry of Education of China, Beijing 100029, China;1. Science and Technology on Thermostructural Composite Materials Laboratory, Northwestern Polytechnical University, Xi’an, Shaanxi 710072, PR China;2. School of Science, Xi''an Polytechnic University, Xi’an, Shaanxi 710048, PR China;3. State Key Laboratory of Solidification Processing, Center for Nano Energy Materials, Northwestern Polytechnical University and Shaanxi Joint Lab of Graphene (NPU), Xi''an 710072, PR China;1. Department of Analytical Chemistry, Medical University of Lublin, Chodźki 4a, 20-093, Lublin, Poland;2. Department of Plant Physiology, Maria Curie-Skłodowska University, Akademicka 19, 20-033 Lublin, Poland;3. Laboratory of Behavioral Research, Medical University of Lublin, Jaczewskiego 8d, 20-090 Lublin, Poland;4. Department of Biology, University of Trnava, Priemyselná 4, 918 43 Trnava, Slovak Republic;5. Department of Pneumology, Oncology and Allergology, Medical University of Lublin, 20-090 Lublin, Poland;1. Department of Chemistry, University of Warsaw, Pasteura 1, 02-093, Warsaw, Poland;2. Institute of Paleobiology, Polish Academy of Sciences, Twarda 51/55, 00-818, Warsaw, Poland;3. National Medicines Institute, Chełmska 30/34, 00-725, Warsaw, Poland;4. Mossakowski Medical Research Centre, Polish Academy of Sciences, Pawińskiego 5, 02-106, Warsaw, Poland;1. ENEA Fusion and Technology for Nuclear Safety and Security Department, Frascati, Rome, Italy;2. RINA Consulting – Centro Sviluppo Materiali S.p.A, Castel Romano, Rome, Italy;1. Department of Electrical Engineering, Korea Advanced Institute of Science and Technology, 291, Daehak-ro, Yusong-gu, Daejeon 305-701, Republic of Korea;2. Hanwha Thales, PGM R&D Center, 491-123, Gyeonggidong-Ro, Chang-Li, Namsa-Myun, Cheoin-Gu, Yongin-City, Gyeonggi-Do 17121, Republic of Korea |
| |
| Abstract: | Genetic algorithm (GA) has a significant effect in the band optimization selection of Partial Least Squares (PLS) correction model. Application of genetic algorithm in selection of characteristic bands can achieve the optimal solution more rapidly, effectively improve measurement accuracy and reduce variables used for modeling. In this study, genetic algorithm as a module conducted band selection for the application of hyperspectral imaging in nondestructive testing of corn seedling leaves, and GA-PLS model was established. In addition, PLS quantitative model of full spectrum and experienced-spectrum region were established in order to suggest the feasibility of genetic algorithm optimizing wave bands, and model robustness was evaluated. There were 12 characteristic bands selected by genetic algorithm. With reflectance values of corn seedling component information at spectral characteristic wavelengths corresponding to 12 characteristic bands as variables, a model about SPAD values of corn leaves acquired was established by PLS, and modeling results showed r = 0.7825. The model results were better than those of PLS model established in full spectrum and experience-based selected bands. The results suggested that genetic algorithm can be used for data optimization and screening before establishing the corn seedling component information model by PLS method and effectively increase measurement accuracy and greatly reduce variables used for modeling. |
| |
| Keywords: | Hyperspectral imaging PLS Corn seeding Chlorophyll content GA |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
