基于计算机视觉的水稻叶片信息研究
|
摘要
植物叶片是植物进行光合作用和蒸腾作用的主要器官,叶片信息直接反映了作物的生长状况,这些信息对于农作物的生长监控具有非常重要的作用。本文以精准农业的发展趋势为指导,应用计算机视觉技术,希望建立更为准确、快捷的叶面积测量方法和利用叶片颜色特征值的特点结合实测的SPAD(Soil Plant Analysis Development)值建立基于计算机视觉的叶片颜色和SPAD值的预测模型。基于以上目的,本论文研究的主要内容包括：对水稻单叶片的面积进行测量,在使用传统的计算机视觉技术测量面积时发现使用叶片背面作为拍照面的测量效果好于使用正面的效果。且在此基础上,探索了两种基于计算机视觉法测量叶片面积的新方法；分析并解决了在大田环境中影响计算机视觉获取图像和测量方法的若干个关键问题,使得在使用计算机视觉对水稻叶片进行SPAD值预测时,精确度更高；在以上结论的基础上分别建立了实验室环境下和大田环境下水稻叶片颜色特征值与SPAD值的模拟模型,并预测其SPAD值,其精度最高可分别达到：97.5%和96.59%。
本文的主要创新点在于：提出了基于计算机视觉的水稻叶片叶色网格分析法,能将水稻叶片表面分割成若干个区域,分析各个区域的颜色特征为使用SPAD仪测量位点选择提供了理论依据。
Leaf is the main organ for plant photosynthesis and transpiration. Blade information directly reflects the growth status of crops, which is very important to monitor the growth of crops. In this paper, the trend of precision agriculture guidance using computer vision technology, hoping to establish a more accurate and efficient measurement of leaf area and use of leaf color characteristic value of the features of SPAD(Soil Plant Analysis Development) values established by actual computer vision-based color and SPAD values of leaves prediction model. Based on the above purpose, the main contents of this paper include:In the process of measuring the rice leaf area, it's better to measure the leaf back than the leaf surface by using traditional computer vision techniques. Based on the above research, new method in measuring rice leaf area of computer vision measurement is explored; To analyze and solve the impact of the environment in field environment computer vision to obtain images and measurements of a number of key issues, making the use of computer vision, SPAD values of rice leaves to predict when the greater accuracy; In the above conclusions were established on the basis of a laboratory environment and field environment, the color characteristics of rice leaf SPAD value and the value of a simulation model and predict the SPAD value, and its accuracy is up to, respectively:97.5% and 96.59%;
The main innovation of this paper is propose the rice leaf color grid analysis based on computer vision, which can divide the rice leaf surface into several regions, and analyze the color characteristics of each region, supplying the theoretical basis to the selection of SPAD sites.
本文的主要创新点在于：提出了基于计算机视觉的水稻叶片叶色网格分析法,能将水稻叶片表面分割成若干个区域,分析各个区域的颜色特征为使用SPAD仪测量位点选择提供了理论依据。
Leaf is the main organ for plant photosynthesis and transpiration. Blade information directly reflects the growth status of crops, which is very important to monitor the growth of crops. In this paper, the trend of precision agriculture guidance using computer vision technology, hoping to establish a more accurate and efficient measurement of leaf area and use of leaf color characteristic value of the features of SPAD(Soil Plant Analysis Development) values established by actual computer vision-based color and SPAD values of leaves prediction model. Based on the above purpose, the main contents of this paper include:In the process of measuring the rice leaf area, it's better to measure the leaf back than the leaf surface by using traditional computer vision techniques. Based on the above research, new method in measuring rice leaf area of computer vision measurement is explored; To analyze and solve the impact of the environment in field environment computer vision to obtain images and measurements of a number of key issues, making the use of computer vision, SPAD values of rice leaves to predict when the greater accuracy; In the above conclusions were established on the basis of a laboratory environment and field environment, the color characteristics of rice leaf SPAD value and the value of a simulation model and predict the SPAD value, and its accuracy is up to, respectively:97.5% and 96.59%;
The main innovation of this paper is propose the rice leaf color grid analysis based on computer vision, which can divide the rice leaf surface into several regions, and analyze the color characteristics of each region, supplying the theoretical basis to the selection of SPAD sites.
引文
[1]Trooien T P,Heermann T F.Measurement and Simulation of Potato Leaf area using image Processing Ⅲ Measurement[J], Transactions of the ASAE,1992,35(5): 1791-1721
[2]Hack G R..The use of image processing under greenhouse conditions for growth and climate control[J].Acta Horticulturae,1993,440,377-382
[3]徐贵力,毛罕平,胡永光.基于计算机视觉技术参考物法测量叶片面积[J].农业工程学报,2002,18(1)：154-158
[4]陈鼎才,王定成,查金水.基于机器视觉的现实叶面积测量方法的研究[J].计算机应用,2006,26(5)：1226-1228
[5]管鹤卿.基于计算机视觉技术的油菜叶片信息研究[D].湖南：湖南农业大学,2007,硕士论文
[6]常丽英,顾东祥,张文宇,等.水稻叶片伸长过程的模拟模型[J].作物学报,2008,34(2)：311-317
[7]刘岩,陆建辉,曹宏鑫,等.基于生物量的水稻叶片主要几何属性模型研究[J].中国农业科学,2009,42(11)：4093-4099
[8]Habib QNadipuram R P,John J E,et al.A neuron-fuzzy approach for insect classification[c].World Automation Congress,Thurd International Symposium on Soft Computing for Industry Maui,Hawaii,2000
[9]陈佳娟,纪寿文,李娟,等.采用计算机视觉进行棉花虫害程度的自动测定[J].农业工程学报,2001,17(2)：157-160
[10]赵玉霞,王克如,白中英,等.基于图像识别的玉米叶部病害诊断研究[J].中国农业科学,2007,40(4)：698-703
[11]刘立波,周国民.基于多层感知神经网络的水稻叶瘟病识别方法[J].农业工程学报,2009,25(增2)：213-217
[12]刘小川,艾矫燕,唐纪良,等.基于自适应模糊阈值的植物黑腐病叶片病斑的分割[J].中国图象图形学报,14(7)：1334-1340
[13]Guyer D E,Miles G E,Gaultney L D,Schreiber M M.Application of machine vision to shape analysis in leaf and plant identification[J].Trans of the ASAE,1993,36(1):163-171
[14]朱静,田兴军,陈彬,等.植物叶形的计算机识别系统[J].植物通报,2005,22(5)：599-604
[15]王晓峰,黄德双,杜吉祥,等.叶片图像特征提取与识别技术的研究[J].计算机工程与应用,2006,42(3)：190-193
[16]郑小东,王晓洁,赵中堂.基于形状特征的植物叶柄与叶片分割算法[J].计算机工程与设计,2010,31(4)：918-921
[17]Singh N,Casady W W,Costello TA.Maehine vision based nitrogen management models for rice[J].Trans of the ASAE,1996,39(3):1899-1904
[18]雷咏雯,危常州,冶军,等.计算机辅助叶色分析进行棉花氮素营养诊断的初步研究[J].石河子大学学报,2004,22(2)：113-116
[19]李小正,谢瑞芝,王克如,等.利用神经网络提取棉花叶片数字图像氮素含量的初步研究[J].作物学报,2007,33(10)：1662-1666
[20]王方永,李少昆,王克如,等.基于机器视觉的棉花群体叶绿素检测[J].作物学报,2007,33(12)：2041-2046
[21]王晓静,张炎,李馨,等.地面数字图像技术在棉花氮素营养诊断中的初步研究[J].棉花学报,2007,19(2)：106-113
[22]Ahmad I S,Reid J F.Evolution of color representations for maize image[J].Agricultural Engineering Research,1996(63):185-196
[23]Gautam,Ramesh Kumar,Panigrahi,et al.Image Processing Techniques and Neural Network Models for Predicting Plant Nitrate Using Aerial Images [J].Proceedings of the Interna-tional Joint.Conference on Neural Networks,2003(2):1031-1036
[24]陈维君,周启发,黄敬峰.用高光谱植被指数估算水稻乳熟期后叶片和穗的色素含量[J].中国水稻科学,2006,20(4)：434-439
[25]张浩,姚旭国,张小斌,等.基于多光谱图像的水稻叶片叶绿素和籽粒氮素含量检测研究[J].中国水稻科学,2008,22(5)：555-558
[26]贾良良,范明生,张福锁,等.应用数码相机进行水稻氮营养诊断[J].2009,8：2176-2179
[27]祝锦霞,邓劲松,石媛媛,等.基于水稻扫描叶片图像特征的氮素营养诊断研究[J].2009(8)：2171-2175
[28]邓启云,袁隆平,蔡义东,等.超级杂交稻模式株型的光合优势[J].作物学报,2006,32(9)：1287-1293
[29]Shimizu H,Heins RD,Computer-vision-based system for plant growth analysis [J]. Trans of the ASAE,1995,38(3):959-964
[30]Caeady W W,Singh N,Costello TA,Machine vision for measurement of rice canopy dimensions[J]. Trans of the ASAE,1996,39(5):1891-1898
[31]Rafael C.Gonzalez, Richard E.Woods. Digital Image Processing[M]. New Jersey: Prentice Hall,2002
[32]Rafael C.Gonzalez, Richard E.Woods, Steven L. Eddins. 数字图像处理 (MATLAB版)[M].北京：电子工业出版社,2006.4
[33]Rafael C. Gonzalez, Richard E. Woods, Steven L. Eddins. Digital Image Processing Using MATLAB [M].New Jersey:Prentice Hall,2004
[34]Hong-sun Yun, Won-ok Lee, Hoon Chung. A Computer Vision System for Rice Kernel Quality Evaluation R Pa per No:023130 An ASAEMeeting Presentation.
[35]Kawamura S, Natsu ga M, Takekura K, et al. Development of automatic rice-quality inspect -ion system[J]. Computers and Electronics in Agriculture,2003,40(1-3):115-126
[36]应义斌,傅宾忠,蒋亦元,等.机器视觉技术在农业生产自动化中的应用[J].农业工程学报,1999,15(3)：199-253
[37]Shuso kawamur, Motoyasu Natsuga. Development of automatic rice-quality inspection system [J]. Computers and Electronics in Agriculture,2003,40(2): 115-126
[38]Tao Y, Chance L, Baoren L. Full scale fruit vision sorting system design-factors and consideration [J].proceedings of the FPAC IV Conference,2003,Illinois: 14-22
[39]蔡鸿昌,崔海信,宋卫堂,等.黄瓜出花期叶片光合色素含量与颜色特征的初步研究[J].农业工程学报,2006,22(9)：34-37
[40]陶密霖,徐光.机器视觉中的颜色信息问题及应用[J].科学通报,2001,46(3)：178-190
[41]林开颜,吴军辉,徐立鸿.彩色图像分割方法综述[J].中国图像图形学报,2005,10(1)：325-326
[42]凌云,王一鸣,孙明,等.基于机器视觉的大米外观品质检测装置[J].农业机械学报,2005,36(9)：89-92
[43]陈建华,姚青,谢绍军,等.机器视觉在稻米粒型检测中的应用[J].中国水稻科学,2007,21：669-672
[44]王沐青,张明普,秦德荣,等.杂交稻汕优63高产的群体特征与促控技术[J].农业科技通讯,1994,11：6-7
[45]叶宏宝,孟亚丽,汤亮,等.水稻叶龄与叶面积指数动态的模拟研究[J].中国水稻科学2008,22(6)：625-630
[46]刘华波,贺立源,马文杰,等.透射图像颜色特征在烟叶识别中的应用探索[J].农业工程学报,2007,22(9)：169-171
[47]应义斌,饶秀勤,马俊福.柑橘成熟度机器视觉无损检测方法研究[J].农业工程学报,2004,20(2)：144-147
[48]林开颜,吴军辉,徐立鸿.基于计算机视觉技术的水果形状分级方法[J].农业机械学报,2005,36(6)：71-74
[49]章海亮,左雪平.苹果自动分级中计算机视觉信息处理技术的研究[J].农机化研究,2006,1：28-30
[50]金晶,廖桂平,李锦卫,等.农产品品质无损检测概述[J].农业网络信息,2008,2：90-93
[51]童钊,廖桂平,李锦卫,等.机器视觉技术在农产品检测中的应用[J].农业网络信息,2008,11：18-21
[52]Suzuki T. Measurement of growth of Plug Seedlings by mage processing in broccoli[J].Acta Hortieulturae,1995(399):333-343
[53]Adamsen F J,Paul J Pinter,J r EM Bamcs,et al.Measuring wheat senescence with a digital eamera[J].Crop Sci,1999(39):719-724
[54]Chien,Chung-Fang,Lin,et al.Non-destructive measurement of Vegetable Seedling Leaf Area Using Elliptical Hough Transform[J].Engineering Solutions for a New Century,2000(2):193-209
[55]Van Henten E J,Bontsema J.Non-destructive crop measurements by image processing for growth control[J]. Journal of Agricultural Engineering Researching, 1995(61):97-105
[56]Ling P P,Cheng Z,Musacchio D J.Quantification of somatic coffee embryo growth using machine vision[J].Transactions of the ASAE,1996:1911-1917
[57]Tao Y,Heinemann P H,Varghese Z,et al.Machine vision for color inspection of potatoes and apples[J].Trans of the ASAE,1995,38(5)2:1555-1561
[58]Noordam,J.C,Otten,G.W, A Color Vision System for High speed sorting of potatoes[J].AGENG Paper NO:00-AE-002,2000
[59]Rafael C.Gonzalez,Richard E.Woods,Steven L.Eddins.Digital Image Processing Using Matlab[M].北京：电子工业出版社,2006：144-154
[60]Rafael C.Gonzalez,Richard E.Woods.Digital Image Processing Second Edition[M].北京：电子工业出版社,2007：228-240
[61]Jinwei Li,Guiping Liao,Zhongbin Ou,Jing Jin.Rapeseed Seeds Classification by Machine Vision.Proceedings of Workshop on Intelligent Information Technology Application (IITA 2007),222-226
[2]Hack G R..The use of image processing under greenhouse conditions for growth and climate control[J].Acta Horticulturae,1993,440,377-382
[3]徐贵力,毛罕平,胡永光.基于计算机视觉技术参考物法测量叶片面积[J].农业工程学报,2002,18(1)：154-158
[4]陈鼎才,王定成,查金水.基于机器视觉的现实叶面积测量方法的研究[J].计算机应用,2006,26(5)：1226-1228
[5]管鹤卿.基于计算机视觉技术的油菜叶片信息研究[D].湖南：湖南农业大学,2007,硕士论文
[6]常丽英,顾东祥,张文宇,等.水稻叶片伸长过程的模拟模型[J].作物学报,2008,34(2)：311-317
[7]刘岩,陆建辉,曹宏鑫,等.基于生物量的水稻叶片主要几何属性模型研究[J].中国农业科学,2009,42(11)：4093-4099
[8]Habib QNadipuram R P,John J E,et al.A neuron-fuzzy approach for insect classification[c].World Automation Congress,Thurd International Symposium on Soft Computing for Industry Maui,Hawaii,2000
[9]陈佳娟,纪寿文,李娟,等.采用计算机视觉进行棉花虫害程度的自动测定[J].农业工程学报,2001,17(2)：157-160
[10]赵玉霞,王克如,白中英,等.基于图像识别的玉米叶部病害诊断研究[J].中国农业科学,2007,40(4)：698-703
[11]刘立波,周国民.基于多层感知神经网络的水稻叶瘟病识别方法[J].农业工程学报,2009,25(增2)：213-217
[12]刘小川,艾矫燕,唐纪良,等.基于自适应模糊阈值的植物黑腐病叶片病斑的分割[J].中国图象图形学报,14(7)：1334-1340
[13]Guyer D E,Miles G E,Gaultney L D,Schreiber M M.Application of machine vision to shape analysis in leaf and plant identification[J].Trans of the ASAE,1993,36(1):163-171
[14]朱静,田兴军,陈彬,等.植物叶形的计算机识别系统[J].植物通报,2005,22(5)：599-604
[15]王晓峰,黄德双,杜吉祥,等.叶片图像特征提取与识别技术的研究[J].计算机工程与应用,2006,42(3)：190-193
[16]郑小东,王晓洁,赵中堂.基于形状特征的植物叶柄与叶片分割算法[J].计算机工程与设计,2010,31(4)：918-921
[17]Singh N,Casady W W,Costello TA.Maehine vision based nitrogen management models for rice[J].Trans of the ASAE,1996,39(3):1899-1904
[18]雷咏雯,危常州,冶军,等.计算机辅助叶色分析进行棉花氮素营养诊断的初步研究[J].石河子大学学报,2004,22(2)：113-116
[19]李小正,谢瑞芝,王克如,等.利用神经网络提取棉花叶片数字图像氮素含量的初步研究[J].作物学报,2007,33(10)：1662-1666
[20]王方永,李少昆,王克如,等.基于机器视觉的棉花群体叶绿素检测[J].作物学报,2007,33(12)：2041-2046
[21]王晓静,张炎,李馨,等.地面数字图像技术在棉花氮素营养诊断中的初步研究[J].棉花学报,2007,19(2)：106-113
[22]Ahmad I S,Reid J F.Evolution of color representations for maize image[J].Agricultural Engineering Research,1996(63):185-196
[23]Gautam,Ramesh Kumar,Panigrahi,et al.Image Processing Techniques and Neural Network Models for Predicting Plant Nitrate Using Aerial Images [J].Proceedings of the Interna-tional Joint.Conference on Neural Networks,2003(2):1031-1036
[24]陈维君,周启发,黄敬峰.用高光谱植被指数估算水稻乳熟期后叶片和穗的色素含量[J].中国水稻科学,2006,20(4)：434-439
[25]张浩,姚旭国,张小斌,等.基于多光谱图像的水稻叶片叶绿素和籽粒氮素含量检测研究[J].中国水稻科学,2008,22(5)：555-558
[26]贾良良,范明生,张福锁,等.应用数码相机进行水稻氮营养诊断[J].2009,8：2176-2179
[27]祝锦霞,邓劲松,石媛媛,等.基于水稻扫描叶片图像特征的氮素营养诊断研究[J].2009(8)：2171-2175
[28]邓启云,袁隆平,蔡义东,等.超级杂交稻模式株型的光合优势[J].作物学报,2006,32(9)：1287-1293
[29]Shimizu H,Heins RD,Computer-vision-based system for plant growth analysis [J]. Trans of the ASAE,1995,38(3):959-964
[30]Caeady W W,Singh N,Costello TA,Machine vision for measurement of rice canopy dimensions[J]. Trans of the ASAE,1996,39(5):1891-1898
[31]Rafael C.Gonzalez, Richard E.Woods. Digital Image Processing[M]. New Jersey: Prentice Hall,2002
[32]Rafael C.Gonzalez, Richard E.Woods, Steven L. Eddins. 数字图像处理 (MATLAB版)[M].北京：电子工业出版社,2006.4
[33]Rafael C. Gonzalez, Richard E. Woods, Steven L. Eddins. Digital Image Processing Using MATLAB [M].New Jersey:Prentice Hall,2004
[34]Hong-sun Yun, Won-ok Lee, Hoon Chung. A Computer Vision System for Rice Kernel Quality Evaluation R Pa per No:023130 An ASAEMeeting Presentation.
[35]Kawamura S, Natsu ga M, Takekura K, et al. Development of automatic rice-quality inspect -ion system[J]. Computers and Electronics in Agriculture,2003,40(1-3):115-126
[36]应义斌,傅宾忠,蒋亦元,等.机器视觉技术在农业生产自动化中的应用[J].农业工程学报,1999,15(3)：199-253
[37]Shuso kawamur, Motoyasu Natsuga. Development of automatic rice-quality inspection system [J]. Computers and Electronics in Agriculture,2003,40(2): 115-126
[38]Tao Y, Chance L, Baoren L. Full scale fruit vision sorting system design-factors and consideration [J].proceedings of the FPAC IV Conference,2003,Illinois: 14-22
[39]蔡鸿昌,崔海信,宋卫堂,等.黄瓜出花期叶片光合色素含量与颜色特征的初步研究[J].农业工程学报,2006,22(9)：34-37
[40]陶密霖,徐光.机器视觉中的颜色信息问题及应用[J].科学通报,2001,46(3)：178-190
[41]林开颜,吴军辉,徐立鸿.彩色图像分割方法综述[J].中国图像图形学报,2005,10(1)：325-326
[42]凌云,王一鸣,孙明,等.基于机器视觉的大米外观品质检测装置[J].农业机械学报,2005,36(9)：89-92
[43]陈建华,姚青,谢绍军,等.机器视觉在稻米粒型检测中的应用[J].中国水稻科学,2007,21：669-672
[44]王沐青,张明普,秦德荣,等.杂交稻汕优63高产的群体特征与促控技术[J].农业科技通讯,1994,11：6-7
[45]叶宏宝,孟亚丽,汤亮,等.水稻叶龄与叶面积指数动态的模拟研究[J].中国水稻科学2008,22(6)：625-630
[46]刘华波,贺立源,马文杰,等.透射图像颜色特征在烟叶识别中的应用探索[J].农业工程学报,2007,22(9)：169-171
[47]应义斌,饶秀勤,马俊福.柑橘成熟度机器视觉无损检测方法研究[J].农业工程学报,2004,20(2)：144-147
[48]林开颜,吴军辉,徐立鸿.基于计算机视觉技术的水果形状分级方法[J].农业机械学报,2005,36(6)：71-74
[49]章海亮,左雪平.苹果自动分级中计算机视觉信息处理技术的研究[J].农机化研究,2006,1：28-30
[50]金晶,廖桂平,李锦卫,等.农产品品质无损检测概述[J].农业网络信息,2008,2：90-93
[51]童钊,廖桂平,李锦卫,等.机器视觉技术在农产品检测中的应用[J].农业网络信息,2008,11：18-21
[52]Suzuki T. Measurement of growth of Plug Seedlings by mage processing in broccoli[J].Acta Hortieulturae,1995(399):333-343
[53]Adamsen F J,Paul J Pinter,J r EM Bamcs,et al.Measuring wheat senescence with a digital eamera[J].Crop Sci,1999(39):719-724
[54]Chien,Chung-Fang,Lin,et al.Non-destructive measurement of Vegetable Seedling Leaf Area Using Elliptical Hough Transform[J].Engineering Solutions for a New Century,2000(2):193-209
[55]Van Henten E J,Bontsema J.Non-destructive crop measurements by image processing for growth control[J]. Journal of Agricultural Engineering Researching, 1995(61):97-105
[56]Ling P P,Cheng Z,Musacchio D J.Quantification of somatic coffee embryo growth using machine vision[J].Transactions of the ASAE,1996:1911-1917
[57]Tao Y,Heinemann P H,Varghese Z,et al.Machine vision for color inspection of potatoes and apples[J].Trans of the ASAE,1995,38(5)2:1555-1561
[58]Noordam,J.C,Otten,G.W, A Color Vision System for High speed sorting of potatoes[J].AGENG Paper NO:00-AE-002,2000
[59]Rafael C.Gonzalez,Richard E.Woods,Steven L.Eddins.Digital Image Processing Using Matlab[M].北京：电子工业出版社,2006：144-154
[60]Rafael C.Gonzalez,Richard E.Woods.Digital Image Processing Second Edition[M].北京：电子工业出版社,2007：228-240
[61]Jinwei Li,Guiping Liao,Zhongbin Ou,Jing Jin.Rapeseed Seeds Classification by Machine Vision.Proceedings of Workshop on Intelligent Information Technology Application (IITA 2007),222-226