基于机器视觉的玉米叶片透射图像特征识别研究

doi:10.3864/j.issn.0578-1752.2014.03.003

摘要/Abstract

摘要： 【目的】建立玉米品种的叶片透射图像特征数据库，研究特征随品种的变化规律，分析各类特征的识别效果，为进一步研究玉米生长期间的机器视觉品种识别提供依据。【方法】以生产中推广的21个常规玉米品种为供试材料，分别采集拔节期、小喇叭口期、大喇叭口期、抽雄开花期4个生育时期的玉米叶片。在灯箱内，采集每一叶片的高画质透射图像，共计420张。基于Matlab R2009a开发了“玉米叶片特征提取与识别软件”，包括图像预处理、特征提取、神经网络识别和阈值选取4个功能模块。依据开发的特征识别平台，对玉米叶片透射图像进行图像预处理和特征提取。提取形态类、颜色类和纹理类共计48个特征，特征数据量共计20 160条。分析48个特征品种间的变异系数，研究玉米叶片透射图像特征随品种的变化规律。建立BP神经网络模型进行综合识别，分析不同时期单特征的识别效果，寻找玉米叶片透射图像中品种区分能力较强的重要特征。进一步分析不同时期3大类特征及其组合的识别效果。【结果】在玉米的4个生育时期，叶片透射图像3类特征品种间的变异系数差异比较明显，颜色类特征变异系数最大，其次是纹理类特征变异系数，形态类特征变异系数最小，并且这种差异随着玉米的生长十分稳定。在玉米的4个生育时期，叶片透射图像48个特征的品种识别率差异比较明显，为9.52%—29.33%。R分量的标准差、短轴长、H分量的标准差、等面圆直径、H分量的平均值、V分量的标准差、B分量的标准差、不变矩6、椭圆度、S分量的平均值、外接凸多边形面积、B分量的平均值、平滑度、S分量的峰度、S分量的标准差的识别率较高，平均识别率在18%以上。单类特征中，颜色类特征识别率最高，平均86.76%；纹理类特征次之，平均为78.05%；形态类特征最低，平均为68.67%。颜色类特征和纹理类特征识别的稳定性较高，纹理类特征识别效果更稳定一些。组合特征中，形态+颜色特征组合识别率最高，平均识别率为92.29%；颜色+纹理类特征组合次之，平均为90.29%；形态+纹理类特征组合识别率最低，但平均识别率也达到了87.43%。在拔节期，3类特征组合的识别率无明显差异，且都在91%以上。在小喇叭口期，颜色+纹理特征组合识别率大幅上升，形态+颜色特征组合识别率小幅下降，形态+纹理特征组合识别率下降较大，但是仍然维持在82%以上。在其他3个时期，形态+纹理特征组合和颜色+纹理特征组合识别率差别不大，并且形态+颜色特征组合识别率相对较高。【结论】研究结果为玉米叶片透射图像特征的研究与应用提供了比较系统全面的数据，为生长期间玉米品种的识别提供了新的方法和量化依据，也必将在其他作物的识别方面发挥很好的借鉴作用。

关键词: 玉米 , 透射图像 , 机器视觉 , 人工神经网络 , 品种识别

Abstract: 【Objective】The purpose of the study was to create database of characteristics from maize leaf transmission images, analyze the rules of characteristics variation with maize varieties and the recognition results of different types of characteristics in order to provide a basis for further research of identifying maize varieties from leaf transmission image of different growth periods based on machine vision. 【Method】 Twenty-one common varieties of maize were selected as the research materials. The maize leaves at jointing stage, small bell stage, large bell stage and tasselling stage were collected. A total of 420 high quality transmission images of maize leaves were taken in lamp box. The software for characteristic extraction and recognition of maize leaves was designed and developed based on Matlab R2009a, which included image preprocessing module, characteristic extraction module, neural network recognition module and threshold selection module. The transmission images of maize leaves at jointing stage, small bell stage, large bell stage and tasselling stage were pre-processed by the software. Then 48 characteristics of color group, shape group and texture group were extracted from transmission images of maize leaf, and a total of 20 160 characteristic data. In order to study the rules of characteinristics variation with maize varieties, the coefficient of variation of 48 characteristics of leaf transmission image among different maize varieties were analyzed. In order to search the important characteristics with strong ability of identifying maize varieties from transmission images of leaves, the Artificial Neural Network was built and the recognition rate of single characteristics from different time were analyzed. In order to study the recognition results, the recognition rates of the three groups of characteristics and the group combinations of characteristics from different time were further analyzed. 【Result】 The results in 4 stages indicated that there were significant differences in the coefficient of variation of 3 groups of characteristics among different maize varieties. The differences were stable with the growth of maize. The coefficient of variation of color group was the highest, then the texture group and the third was the shape group. The results at 4 stages also indicated that there were significant differences among the recognition rates of 48 single characteristics. The recognition rates were between 9.52% and 29.33%. According to the recognition rates, the important characteristics were in the following order: the standard deviation of R, the minor axis length, the standard deviation of H, the diameter, the average value of H, the standard deviation of V, the standard deviation of B, the invariant moment 6, the eccentricity, the average value of S, the external convex polygon area, the average value of B, the smoothness, the kurtosis of V, and the standard deviation of S. Those average recognition rates was over 18%. The average recognition rate of the color group was the highest which was 86.76%, then the texture group which was 78.05%, and the third was the shape group which was 68.67%. The stability of recognition rates of the texture group was the highest, then the color group and the third was the shape group. The average recognition rate of combinations of shape group and color group was the highest which was 92.29%, then the combinations of color group and texture group which was 90.29%, and the third was the combinations of shape group and texture group which was 87.43%. At jointing stage, there was no obvious difference among the recognition rates of combinations of 3 groups. At small bell stage, the recognition rate of the combinations of color group and texture group rose sharply, while the recognition rate of the combinations of shape group and color group fell slightly, and the recognition rate of the combinations of shape group and texture group showed a marked decline, which still remained above 82%. There was no significant difference between recognition rates of the shape group and texture group and the color group and texture group at jointing stage, large bell stage and tasselling stage. And the recognition rate of combinations of the shape group and color group was always the highest. 【Conclusion】 This study has provided rich primary data on characteristics from maize leaf transmission images, and has provided a quantitative basis and a new method for further research of identifying maize varieties in different growth periods, which will play a good reference role in crop varieties identification.

Key words: maize , transmission image , machine vision , artificial neural network (ANN) , variety identification

唐俊1, 邓立苗2, 陈辉1, 栾涛1, 马文杰1. 基于机器视觉的玉米叶片透射图像特征识别研究[J]. 中国农业科学, 2014, 47(3): 431-440.

TANG Jun-1, DENG Li-Miao-2, CHEN Hui-1, LUAN Tao-1, MA Wen-Jie-1. Research on Maize Leaf Recognition of Characteristics from Transmission Image Based on Machine Vision[J]. Scientia Agricultura Sinica, 2014, 47(3): 431-440.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://www.chinaagrisci.com/CN/10.3864/j.issn.0578-1752.2014.03.003

https://www.chinaagrisci.com/CN/Y2014/V47/I3/431

参考文献

[1]郝建平, 杨锦忠, 杜天庆, 崔福柱, 桑素平. 基于图像处理的玉米品种的种子形态分析及其分类研究. 中国农业科学, 2008, 41(4): 994-1002.

Hao J P, Yang J Z, Du T Q, Cui F Z, Sang S P. A study on basic morphologic information and classification of maize cultivars based on seed image process. Scientia Agricultura Sinica, 2008, 41(4): 994-1002. (in Chinese)

[2]王玉亮, 刘贤喜, 苏庆堂, 王朝娜. 多对象特征提取和优化神经网络的玉米种子品种识别. 农业工程学报, 2010, 26(6) : 199-204.

Wang Y L, Liu X X, Su Q T, Wang Z N. Maize seeds varieties identification based on multi-object feature extraction and optimized neural network. Transactions of the CSAE, 2010, 26(6): 199-204. (in Chinese)

[3]权龙哲, 祝荣欣, 雷溥, 韩豹. 基于K-L变换与LS-SVM的玉米品种识别方法. 农业机械学报, 2010, 41(4): 168-172.

Quan Z L, Zhu R X, Lei B, Han B. Recognition method of maize cultivars based on K-L transform and LS-SVM. Transactions of the Chinese Society for Agricultural Machinery, 2010, 41(4): 168-172. (in Chinese)

[4]Chen X, Xun Y, Li W. Combining discriminant analysis and neural networks for corn variety identification. Computers and Electronics in Agriculture, 2010, 71(S1) : 48-53.

[5]韩仲志, 杨锦忠. 多类支持向量机在玉米品种识别中的应用. 农机化研究, 2011(11) : 159-163.

Han Z Z, Yang J Z. Using multi-variable SVM arithmetic in maize cultivars classifications. Journal of Agricultural Mechanization Research, 2011(11): 159-163. (in Chinese)

[6]孙钟雷, 万鹏, 龙长江. 基于遗传组合网络的玉米品种识别方法. 农机化研究, 2012(8): 102-106.

Sun Z L, Wan P, Long C J. Composite GA－RBF networks for corn variety identification. Journal of Agricultural Mechanization Research, 2012(8): 102-106. (in Chinese)

[7]韩仲志, 杨锦忠, 李言照. 玉米品种图像识别中的影响因素研究. 中国粮油学报, 2012, 27(10): 98-103.

Han Z Z, Yang J Z, Li Y Z. Study on the influencing factors of corn cultivars by image classification. Journal of the Chinese Cereals and Oils Association, 2012, 27(10): 98-103. (in Chinese)

[8]杨锦忠, 张洪生, 赵延明, 宋希云, 王新勤. 玉米穗粒重与果穗三维几何特征关系的定量研究. 中国农业科学, 2010, 43(21): 4367-4374.

Yang J Z, Zhang H S, Zhao Y M, Song X Y, Wang X Q. Quantitative study on the relationships between grain yield and ear 3-D geometry in maize. Scientia Agricultura Sinica, 2010, 43(21): 4367-4374. (in Chinese)

[9]王慧慧, 孙永海, 张贵林, 张婷婷, 李义, 许秀颖. 基于压力和图像的鲜玉米果穗成熟度分级方法. 农业工程学报, 2010, 26(7): 369-373.

Wang H H, Sun Y H, Zhang G L, Zhang T T, Li Y, Xu X Y. Grading method of fresh corn ear maturity based on pressure and image. Transactions of the CSAE, 2010, 26(7): 369-373. (in Chinese)

[10]韩仲志, 杨锦忠. 计数玉米穗行数的机器视觉研究. 玉米科学, 2010, 18(2): 146-148, 152.

Han Z Z, Yang J Z. Vision research on the machine of counting ear rows in maize. Journal of Maize Sciences, 2010, 18(2): 146-148, 152. (in Chinese)

[11]杨锦忠, 张洪生, 郝建平, 杜天庆, 崔福柱, 李娜娜, 梁改梅. 玉米果穗图像单一特征的品种鉴别力评价. 农业工程学报, 2011, 27(1): 196-200.

Yang J Z, Zhang H S, Hao J P, Du T Q, Cui F Z, Li N N, Liang G M. Identifying maize cultivars by single characteristics of ears using image analysis. Transactions of the CSAE, 2011, 27(1): 196-200. (in Chinese)

[12]吴兰兰, 刘俭英, 文友先. 基于分形维数的玉米和杂草图像识别. 农业机械学报, 2009, 40(3): 176-179.

Wu L L, Liu J Y, Wen Y X. Image identification of corn and weed based on fractal dimension. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(3): 176-179. (in Chinese)

[13]侯晨伟, 陈丽. 基于概率神经网络的玉米苗期杂草识别方法的研究. 农机化研究, 2010, 32(11): 41-43, 47.

Hou C W, Chen L. Modeling and simulation analysis of corn stalks stacking mechanism of the corn harvestor. Journal of Agricultural Mechanization Research, 2010, 32(11): 41-43, 47. (in Chinese)

[14]Alberto T, Gonzalo P, Xavier P. A computer vision approach for weeds identification through Support Vector Machines. Applied Soft Computing, 2011, 11(1): 908-915.

[15]Sanyala P, Patel S C. Pattern recognition method to detect two diseases in rice plants. The Imaging Science Journal, 2008, 56(6): 319-325.

[16]王娜, 王克如, 谢瑞芝, 赖军臣, 明博, 李少昆. 基于Fisher判别分析的玉米叶部病害图像识别. 中国农业科学, 2009, 42(11): 3836-3842.

Wang N, Wang K R, Xie R Z, Lai J C, Ming B, Li S K. Maize leaf disease identification based on fisher discrimination analysis. Scientia Agricultura Sinica, 2009, 42(11): 3836-3842. (in Chinese)

[17]马晓丹, 关海鸥, 祁广云, 左豫虎, 金宝石. 植物叶片智能分析系统的设计. 中国农业大学学报, 2012, 17(5): 155-159.

Ma X D, Guan H O, Qi G Y, Zuo Y H, Jin B S. Design on intelligent analysis system of plant-leaf. Journal of China Agricultural University, 2012, 17(5): 155-159. (in Chinese)

[18]宁纪锋, 何东健, 杨蜀秦. 玉米籽粒的尖端和胚部的计算机视觉识别. 农业工程学报, 2004, 20(3): 117-119.

Ning J F, He D J, Yang S Q. Identification of tip cap and germ surface of corn kernel using computer vision. Transactions of the CSAE, 2004, 20(3): 117-119. (in Chinese)

[19]韩仲志, 赵友刚, 杨锦忠. 基于籽粒RGB图像独立分量的玉米胚部特征检测. 农业工程学报, 2010, 26(3): 222-226.

Han Z Z, Zhao Y G, Yang J Z. Detection of embryo based on independent components for kernel RGB images in maize. Transactions of the CSAE, 2010, 26(3): 222-226. (in Chinese)

[20]王传宇, 郭新宇, 温维亮, 苗腾. 基于计算机视觉的玉米籽粒形态测量. 农机化研究, 2011, 33(6): 141-144.

Wang C Y, Guo X Y, Wen W L, Miao T. The measurement of maize kernel shape based on computer vision. Journal of Agricultural Mechanization Research, 2011, 33(6): 141-144. (in Chinese)

[21]万鹏, 孙钟雷, 宗力. 基于计算机视觉的玉米粒形检测方法. 中国粮油学报, 201l, 26(5): 107- 110.

Wan P, Sun Z L, Zong L. Detection method of maize mernel shape based on computer vision. Journal of the Chinese Cereals and Oils Association, 201l, 26(5): 107-110. (in Chinese)

[22]Bruno O M, Plotze R O, Falvo M. Fractal dimension applied to plant identification. Information Sciences, 2008, l78(12): 2722-2733.

[23]张善文, 王献峰. 基于加权局部线性嵌入的植物叶片图像识别方法. 农业工程学报, 2011, 27(12): 141-145.

Zhang S W, Wang X F. Method of plant leaf recognition based on weighted locally linear embedding. Transactions of the CSAE, 2011, 27(12): 141-145. (in Chinese)

[24]王艳菲, 朱俊平, 蔡骋. 基于CENTRIST的植物叶片识别研究与实现. 计算机工程与设计, 2012, 33(11): 4343-4346.

Wang Y F, Zhu J P, Cai C. Research and implementation of plant leaf recognition based on CENTRIST. Computer Engineering and Design, 2012, 33(11): 4343-4346. (in Chinese)

[25]阎庆, 梁栋, 张晶晶. 基于Fisher变换的植物叶片图像识别监督LLE算法. 农业机械学报, 2012, 43(9): 179-183.

Yan Q, Liang D, Zhang J J. Recognition method of plant leaves based on Fisher projection-supervised LLE algorithm. Transactions of the Chinese Society for Agricultural Machinery, 2012, 43(9): 179-183. (in Chinese)

[26]张长胜, 欧阳丹彤, 岳娜, 张永刚. 一种基于遗传算法和 LM 算法的混合学习算法. 吉林大学学报: 理学版, 2008, 46(4): 675-680.

Zhang C S, Ouyang D T, Yue N, Zhang Y G. A hybrid algorithm based on genetic algorithm and levenberg-marquardt. Journal of Jilin University: Science Edition, 2008, 46(4): 675-680. (in Chinese)

[27]梁毅, 刘世洪. 基于遗传算法优化的 BP 神经网络的组合预测模型方法研究. 中国农业科学, 2012, 45(23): 4924-4930.

Liang Y, Liu S H. Research on the combined forecast model method based on BP neural network improved by genetic algorithm. Scientia Agricultura Sinica, 2012, 45(23): 4924-4930. (in Chinese)