用户名: 密码: 验证码:
玉米品种的计算机视觉识别研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
计算机视觉自动检测种子资源品质,是实现农业生产自动化发展的必然趋势。本研究在国内外相关研究基础上,以目前尚未进行的玉米品种识别作为研究方向,选取四个玉米品种作为研究对象,以计算机视觉和模式识别理论为基础,通过获取玉米种子图像,对图像处理、分析,提取反映玉米品种形态结构的特征参数,用人工神经网络方法识别玉米品种。
     主要研究内容如下:
     1、根据计算机视觉检测玉米品种要求,建立玉米种子图像获取和分析的计算机视觉系统。
     2、获取玉米种子图像,对图像进行背景分割,去除噪音,提取单个种子。用边界跟踪法定位图像中一粒玉米种子边界,用种子填充法计算玉米种子在图像中的区域,保存该区域,用于下一步处理。
     3、分析玉米种子的形态结构,定义了一组基本参数,将反映玉米品种形态结构的特征参数分为三组,分别是颜色参数、形状参数、大小参数。选用HLS颜色模型来表示颜色特征。
     4、研究并提出玉米种子尖端位置检测算法,通过分析玉米种子尖端位置结构特征,用局部最大曲率法检测玉米种子尖端位置,试验表明,该算法准确率高,为正确计算玉米种子特征参数奠定了基础。
     5、研究并提出了判别玉米种子胚乳部所在面的检测算法,该算法利用胚乳部所在面的图像特征。试验表明,提出的算法能够正确识别所拍摄玉米种子图像是否是胚乳部所在面。
     6、对一粒玉米种子两面各拍一幅图像,研究两面特征,分析它们之间差异及对品种特征的影响。
     7、用三层BP神经网络识别玉米品种,通过训练确定网络参数,优化组合输入参数,识别率为93%。
     8、基于Visual C++6.0,开发了一套玉米品种计算机视觉识别研究软件,该软件界面友好,功能完备。能够完成图像的背景分割、平滑、单个玉米种子提取,种子特征参数的计算,生成特征参数数据文件及二值化、直方图运算。
Automation of agriculture producing is the future trend by computer vision inspects the quality of seed. There has made much research on the quality of seed. Select four different corn breed and research the breed identification of corn seed by computer vision. On the basis of pattern recognition analyze image and extract feature parameters, which reflect the morphology feature of corn breed. In the research using the neural network identifies four different corn breeds.
    The research mainly includes eight parts:
    1. A computer vision system was developed to acquire the image of corn seed and detect the seed breed.
    2. The image of corn seeds which are disconnected is acquired is processed including background segment and smoothing image and extract single corn seed. The region of a corn seed is located by boundary following and seed filling and saved as a new image file.
    3. The morphological structure of corn seed is analyzed. Three group parameters including color and shape and size show the morphological structure of corn seed. Color Parameters is denoted by HLS color mode.
    4. Developing a algorithm which looks for the tip of corn seed by the structure feature of tip. The result shows that it is accurate to use local max curvature to locate its tip.
    5. Developing is an algorithm that distinguishes the face of endosperm by the image feature of endosperm. The result shows it can identify effectively the face of endosperm.
    6. Take two surface pictures of corn seed and research feature parameters of two surfaces and their influence.
    7 The corn breed is identified by three layer BP neural network. To optimize the network structure and input vector, the result shows the method can identify corn breed to 93%.
    8 A software system that the corn breed is identified by computer vision is developed. The software has friend interface and stronger function, which can finish background segment and smooth of the image and extract single corn seed and figure out the feature parameter.
引文
[1] Ahmad I S.Reid J F.Evaluation of Color Representations for Maize Image.Journal of Agricultural Engineering Research, 1996, 63: 185-196
    [2] Berlage A G, Cooper T M, Aristazabal J F.Machine vision identification of diploid and teraploid tyegrass seed.Transaction of the ASAE. 1988, Vol.31 (1) : 24-27
    [3] Berlow S N, D J Aneshansley, J A Throop, et al.Computer analysis of ultrasonic images for grading beef, 1989, ASAE Paper No.89-6569
    [4] B.Ni, M.R.Paulsen 等.Corn kernel crown shape identification using image processing.Transactions of ASAE, 1996, Vol.40 (3) : 833-838.
    [5] B.Ni, M.R.Paulsen 等.Design of an Automated Corn Kernel Inspection System For Machine Vision.Transactions of the ASAE.1997, Vol.40 (2) : 491-497
    [6] Das K, M D Evans.Detecting fertility of hatching eggs using machine vision II: neural network classifiers.Trans of-the ASAE, 1992, 35 (6) : 2035-2041
    [7] Ding K, Gunasekaran S.Shape feature extraction and classification of food material using computer vision.Transactions of ASAE, 1994, 37 (5) : 1537-1545.
    [8] Ding, K, R.V.Morey 等.Corn quality evaluation with computer vision. ASAE Paper NO.90-3532. St Joseph, MI: ASAE
    [9] D.Wang, F.E.Dowel 1, R.E.Lacey.Single Wheat Kernel Color Classification Using Neural Network.Transactions of The ASAE. 1999, Vol.40 (1) : 233-240
    [10] Gunasekaran S, Cooper T M 等.Image Processing for stress cracks in corn kernels.Transactions of the ASAE, 1988, Vol 31(1) : 257-263
    [11] Lai,F.s.L.Zayas 等.1986. Application of pattern recognition techniques in the analysis of cereal grains.Cereal Chemistry 63(2) : 168-172
    [12] J.Liu, M.R.Paulsen.Corn Whiteness measurement and classification using machine vision.Transaction of the ASAE.2000, Vol 43(3) : 757-763
    [13] J.Pal iwal, N.S.Shashidhar.Grain kernel identification using kernel signature.Transactions of the ASAE.1999, Vol: 42 (6) : 1921-1924
    [14] K.Liao, M.R.Paulsen.Corn kernel breakage classification by machine vision using a neural network classifier.1993, Vol 36 (6) : 1949-1953
    [15] Liao K, Paulsen MR, Reid JF et al.Corn kernel shape identification by machine vision using a neural network classifier.ASAE Paper, No.927017,1992.
    [16] Liao K, Reid J J, Ni B, et al.Corn kernel shape identification by machine vision using a neural network classifier.ASAE Paper, 1992. 92-105
    
    
    [17] Liao K, Paul sen M R, Reid J F.Real-time detection of color and surface defects of maize kernels using machine vision.Jounal of Agricultural Engineering Research, 1994, 59: 263-271
    [18] Miller B K, MJ Cooper, A G Berlage, et al.Image processing for stress cracks in corn kernels.Trans of the ASAE.1987, 30(1) :266-271
    [19] Panigrahi S, Misra M.Background segmentation and dimensional measurement of corn germplasm.Trans of the ASAE, 1995, Vol.38(1) : 291-297
    [20] Paulsen MR, W F Me Clure.Illumination for computer vision system. Trans of the ASAE.1989, ASAE Paper No.89-3569
    [21] P.Shatadal,D.S.Jayas.Digital image analysis for software separation and classification of touching grains: I.disconnect algorithm.Transactions of the ASAE.1995, Vol: 38 (2) : 635-643
    [22] P.Shatadal,D.S.Jayas.Digital image analysis for software separation and classification of touching grains : II.classification. Transactions of the ASAE.1995, Vol: 38 (2) : 635-643
    [23] Seginer I, Elster R T, Goodrum J.Wet al Plant wilt detection by computer-vision tracking of leaf t ips.Trans of the ASAE, 1992,35(5) : 1563-1567
    [24] Shatadal-P, Jayas-Ds.Seed classification using machine vis ion.Canadian Agricultural Engineer ing, 1995, 373 (3) : 163-167
    [25] Slaughter-DC, Chen-P.Computer Vision guidance system for precis ion cultivation.ASAE Annual International Meeting, Minneapolis, Minnesota, USA, 1997: 10-14
    [26] Tanaka M.Development of an Image database for forests.Journal of Forest Planning., 1996, 2 (1) : 51-56
    [27] Tao Y.Closed-loop search method for on-line automatic calibrations of multi-camera inspection systems.Trans of the ASAE. 1998a, 41 (5) : 1549-1555
    [28] Tao Y, Heinemann .P, Varghese .Zet al.Machine vision for color inspection of potatoes and apples.Transactions of ASAE, 1995,38(5) :1555"1561.
    [29] Tao Y, L Chance, L Baoren.Full scale fruit vision sorting system design-factors and considerations.Proceedings of the FPAC IV Conference, 1995, Chicago, Illinois, 14-22
    [30] Taylor R W., Rehkugler G E, Throop J A.Apple bruise detection using a digital line scan camera system.In Agricultural electronics-1983 and Beyond-Volume II.St.Joseph, ASAE, 1984
    [31] Throop J A, G E Rehkugler, and B L Upchurch.Application of computer vision for detecting water core in apples.Trans of the ASAE, 1989, 32(6) : 2087-2092
    [32] Tollner-EW, Shahin-MA.ASAE CASE SCGR.Annual Internalional Meeting, Toronto.Ontario.Canada, 1999, 19pp
    [33] Vigneault-C, Roger-C.Study of the behavior of insects with the
    
    assistance of compute vision. Resume-des-Recherches Centre-de recherche et development en horticulture, Saint Jean sur richelieu,-quebec, 1994,23:52-53
    [34]W.D. Batchelor, J.W. Jones. Comparisons of Methods to Compute Peanut Seed Size Distribution By Crop Growth Models. Vol. 39(2):737-744
    [35]Wilhoit JH, Kutz Lj. Machine vision system for quality control assessment of bare root pine seedlings. Southern-Journal of applied forestry, 1997,21(2): 90-96
    [36]Y.C Chang, J.F. Reid. Characterization of Color Vision System. Transactions of the ASAE. 1996, Vol. 39(1):263-273
    [37]Y. Chtioui, S. Panigraphi. Rough Sets Theory as a Pattern Classification Tool For Quality Assessment of Edible. Transactions of the ASAE. 1999, Vol. 42(4):1145-1152
    [38]Zayas I, Converse H, Steele J. Discrimination of whole from broken corn kernels with image analysis. Trans of the ASAE, 1990, Vol. 33(5):1642-1646
    [39]Zayas, L, F.s. Lai z. 1986. Discrimination between wheat classes and varieties by image analysis. Cereal Chemistry 63(1):52-56
    [40]Zayas, L, H. Converse. 1990. Discrimination of whole from broken corn kernels with image analysis. Transactions of ASAE 33(5):1642-1646
    [41]Zayas, L, Y. Pomeranz. Lai. 1989. Discrimination of wheat and nonwheat components in grain samples by image analysis. Cereal Chemistry 66(33):233-237
    [42]Zayas, L, Y. Pomeranz. Lai 1985. Discrimination between Arthur and Arkan wheats by image analysis-Note. Cereal Chemistry 62(2):478-480.
    [43]Zhang N, Chaisattapagon C. Effective criteria for weed identification in wheat fields using machine vision. Trans of the ASAE, 1995,38(3): 965-974
    [44]Zhang-ShuHai, Takahashi-T. Studies on automation of work in orchards (part1). Detection of apple by pattern recognition. Journal of the Japanese Society of Agricultural Machinery, 1996,58:1,9-16
    [45]Zion-B, Shklyar-A. Computers-and-Electronics-in-Agriculture, 1999,23(3): 175-187
    [46]何东健,杨青等.果实表面颜色计算机视觉分级技术的研究.农业工程学报,1998,14(3):202-204
    [47]何东健,杨青等.果实缺陷面积的计算机视觉测定研究.农业工程学报,1997,13(4):156-160
    [48]何东健,杨青等.用人工神经网络进行果实颜色分级技术研究,西北农业大学学报,1998,26(6):109-112
    [49]何东健,耿楠.用活动边界模型精确检测果实表面缺陷.农业工程学报,2001,17(5):159-162
    [50]何东健,杨青著.实用图像处理技术.西安:陕西科技出版社,1998
    [51]何东健,杨青,薛少平.农业机器人视觉系统及识别技术研究.中国农业大学学报,1996(第一卷,增刊),38-43
    
    
    [52]何东健,杨青,熊运章,薛少平.农产品近红外分光反射特性及近红外图像处理在农业中的应用.农业工程学报,1996,12(4):80-40
    [53]何东健,前川昭孝,森岛博.苹果内部品质在线近红外分光检测装置及试验.2001,17(1):146-148
    [54]耿楠,党革荣.判别分析法确定最佳阈值的快速算法.西北农林科技大学学报,2001,29(6):119-121
    [55]耿楠,何东健等.小麦生长信息计算机视觉检测技术研究.农业工程学报.2001.17(1):136-139
    [56]龙满生,宁纪锋.基于遗传神经网络的苹果综合分级系统.西北农林科技大学学报.2001,29(6):119-121
    [57]宁纪锋,龙满生等.农业领域中的计算机视觉研究.计算机与农业.2001,1:1-3
    [58]宁纪锋,何东健等.基于图像形态学的球形果实品质检测研究.农机化研究.2001.3:28-30
    [59]应义斌,景寒松等.黄花梨果形的机器视觉识别方法研究.农业工程学报,1999,15(1):192-196
    [60]应义斌,铙秀勤等.机器视觉技术在农产品品质自动识别中的应用(Ⅰ).农业工程学报,2000,16(1):103-108
    [61]应义斌,铙秀勤等.机器视觉技术在农产品品质自动识别中的应用.农业工程学报,2000,16(3):4-8
    [62]应义斌,章文英,蒋亦元等.机器视觉在农产品收获和加工自动化中的应用.农业工程学报,2000,31(3):112-114
    [63]应义斌.水果图像的背景分割和边缘检测技术研究.浙江大学学报(农业与生命科学版),2000,25(1):230-232
    [64]李庆中,汪懋华.基于计算机视觉的苹果自动分级系统硬件开发.农业机械学报,2000,31(2):56-59
    [65]邓继中,张秦岭,洪添胜.一种用于水果碰压伤面积检测的数学模型.华南农业大学学报,2000,23(2):85-87
    [66]张秦岭,邓继忠.应用计算机视觉技术对梨碰压伤的检测.农业工程学报.1999,15(1):205-209
    [67]刘成良,王红永等.秧苗嫁接机器人视觉与识别的研究.农业机械学报.2001,32(4);38-41
    [68]张书慧,陈晓光等.苹果、桃等农副产品品质检测与分级图像处理系统的研究.农业工程学报.1999,15(1):201-204
    [69]董美对,何勇等.精确农业-21世纪的农业工程技术.浙江大学学报(农业与生命科学版),2000,26(4):433-436
    [70]杨邦杰,裴志远.农作物长势的定义与遥感监测.农业工程学报,1999,15(3):214-218
    [71]周云山,李强等.计算机视觉在蘑菇采摘机器上的应用.农业工程学报,1995,11(4):27-32
    [72]陈晓光,于海业等.应用图像处理技术进行蔬菜苗特征量识别.农业工程学报,1995,11(4):21-26
    [73]宋韬,曾德超.基于人工神经网络的玉米籽粒形态识别方法研究.农业工程学报,1996.12(1):177-181.
    
    
    [74]吴羡,白景峰等.针叶树种计算机视觉苗木自动分级系统的研究.东北林业大学学报,1998,26(4):32-35
    [75]王金星,张守娟等.长白落叶松人工林木材构造计算机视分析的研究.东北林业大学学报,1995,23(3):20-25.
    [76]陈顺山,汪懋华,谭梅芳.奶牛体形图像信息系统研究.农业工程学报,1996.12(3):149-152.
    [77]徐娟,汪懋华.图像形态学在苹果自动分级视觉信息处理中果梗判别与边缘检测中的应用.农业工程学报,1999,15(2):177-180
    [78]中国农业标准汇编(经济作物卷).北京:中国标准出版社,1997:281-294
    [79]崔屹.图像处理与分析-数学形态学方法及应用.北京:科学出版社,2000:15-42
    [80]向世英,贾希海等.真假种子的判别.北京:中国计量出版社,1999
    [81]张弓,D.S.Jayas.谷物纹理特征的识别.农业工程学报.2001,Vol:12(1):149-153
    [82]胡少兴,马成林等.排种器性能检测中种子位置智能检测方法.农业机械学报》.2001,Vol:32(3):36-38
    [83]周洪生.玉米种子大全.北京:中国农业出版社,2000
    [84]向世英,贾希海等.真假种子的判别.北京:中国计量出版社,1999
    [85]孙致良,杨国枝.实用种子检验技术.北京:中国农业出版社,1993
    [86]浙江农业大学种子教研组编.农作物种子学.杭州:浙江科学技术出版社,1981
    [87]李家义,支巨振,黄亚军等编,1993国际种子检验规程.上海:上海科学技术出版社
    [88]GB/T3543.1-3543.7-1995.农作物种子检验规程.国家技术监督局发布.北京:中国农业出版社
    [89]吕凤军.数字图像处理编程入门.北京:清华大学出版社.1999
    [90]章毓晋.图像工程(上册)-图像处理和分析.北京:清华大学出版社,1999
    [91]章毓晋.图像工程(下册)-图像理解和计算视觉.北京:清华大学出版社,2000
    [92]贾云得.机器视觉.贾云得.北京:科学出版社.2000
    [93]周孝宽,曹晓光等.实用微机图像处理.北京:航空航天大学出版社
    [94]客观澳.计算机图像处理.北京:清华大学出版社
    [95]崔屹.图像处理与分析-数学形态学方法及应用.北京:科学出版社,2000
    [96]孙家广,杨长贵.计算机图形学.北京:清华大学出版社,1996
    [97]David J.Kruglinski等.Vlsual C++技术内幕(第四版).北京:清华大学出版社:1999
    [98]乔森,杨志刚.Visual C++高级编程技术(多媒体篇).北京:中国铁道出版社,2000
    [99]周长发.精通Visual C++图像编程.北京:电工子工业出版社:2000
    [100]Kate Gregory.Visual C++6.0使用手册.北京:机械工业出版社:1999
    [101]何斌,马天予.Visual C++数字图像处理.北京:中国邮电出版社,2001
    [102]Visual C++6.0入门与提高
    [103]丛爽.面向MATLAB工具箱的神经网络理论与应用.合肥:中国科学技术大学出版社:1998
    
    
    [104]边肇祺,张学工.模式识别.北京:清华大学出版社:2000
    [105]王晓丹,吴崇明.基于NATLAB 的系统分析与处理—图像处理.西安:西安电子科技大学出版社:2000
    [106]许志祥,卢宏,沈剑.摄像机标定及其误差分析.自动化学报.1993,19(1):114-117
    [107]钟瑞永,李芳繁.使用双眼立体机器视觉定位果树.[台]农业工程学报,1993,39(4):74-87
    [108]大恒图像视觉有限公司.VIDEO-PCI-C用户使用手册.1996.8.
    [109]李在清等.颜色测量基础.北京:北京技术标准出版社,1980
    [110]王丰元等.计算机视觉摄影测量三维控制场设计.农业机械学报.1998,29(3):158~160

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700