基于计算机视觉的水稻叶部病害识别研究

doi:10.3864/j.issn.0578-1752.2014.04.006

摘要/Abstract

摘要： 【目的】文章重点分析了病健交界特征参数、病害识别流程对提高病害识别准确率的影响。实现水稻叶部15种主要病害的准确识别，尤其是相似病害的判断。【方法】（1）病斑图像获取：水稻叶部病害图像来源包括水稻大田、病害图册和病害数据库，文中选用改进的mean shift图像分割算法提取病叶图像中的病斑并根据相关方程获取病斑特征信息。（2）特征参数的选择与设计：首先选取一至三阶颜色矩和颜色直方图作为病害的颜色特征参数，选取球状性、偏心率和不变矩作为病斑的形状特征参数，选取角二阶矩、对比度和相关作为病斑的纹理特征参数；然后针对相似病斑误报率高的问题提出一种病健交界特征参数，通过病斑内部、边缘和外围颜色上的差异描述该特征，并根据3个区域相互间归一化颜色直方图的欧氏距离计算该项特征参数，该参数可以用于描述病斑与健康部分交界处的特征。（3）病害识别流程的设计：根据病害在颜色、形状、纹理、病健交界4个特征上差异的显著程度设计完成病害识别流程，流程中首先通过颜色特征识别病害，对于通过颜色特征无法识别的病害再通过形态特征识别，倘若形态特征依然无法识别则通过纹理和病健交界特征进行最终识别。（4）病害识别模型的建立：将病害数据分成两部分，一部分用于建立模型，另一部分用于模型的验证；利用LibSVM程序包完成建模，其中svmtrain函数用于建立支持向量机模型，Grid程序用于优化参数，svmpredict函数用于对模型进行验证。【结果】15种水稻叶部病斑可以从复杂的背景中分割出来，并可快速准确的被识别，平均识别准确率为92.67%，平均漏报率为7.00%，最大漏报率和误报率分别为15.00%和25.00%；病健交界特征参数引入后，识别准确率提高了14.00%，平均漏报率降低了7.50%，漏报率最大降幅为20.00%，误报率最大降幅为65.00%；与用所有特征参数直接进行病害识别相比，采用本文提出的识别流程进行病害识别的准确率提高了12.67%，漏报率降低了9.33%，一些病害的漏报率和误报率降幅达30.00%以上；在识别流程各步骤中，颜色特征识别环节的平均准确率为96.71%，漏报率和误报率均未超过10.00%；形态特征识别环节的平均准确率为94.17%，漏报率和误报率均未超过15.00%；纹理和病健交界特征识别环节的平均准确率为91.50%，漏报率和误报率均未超过25.00%。【结论】利用mean shift图像分割算法可以准确分割水稻叶部病斑；基于支持向量机模型的分类方法可以对15种水稻病斑准确分类；论文中提出的病健交界特征参数以及病斑识别流程均提高了病斑的识别准确率；病健交界特征参数对提高一些相似病害的识别精度效果显著；将这些方法相结合可以有效对水稻常见叶部病害进行识别，为水稻病害的田间智能诊断提供技术支撑。

关键词: 水稻病害 , 计算机视觉 , 图像处理 , 病健交界 , 均值漂移 , 支持向量机

Abstract: 【Objective】The purpose of this article was to recognize 15 kinds of rice disease located at leaf accurately, especially among the similar diseases. 【Method】 For disease spot image acquisition, the original image was got by digital camera from paddy fields and modified mean shift image segmentation algorithm was used to extract rice disease spot from the original image. For the selection and design of feature parameters, Firstly, the color moments arranged from first to third order and color histogram were selected as color feature parameters, and sphericity, eccentricity, invariant moment as shape feature parameters, and angular second moment, contrast, correlation as texture feature parameters. Secondly, junction feature parameters were designed aiming at high missing report rate of similar diseases, this feature was established based on the difference between inside, margin and periphery of disease spot, and euclidean distances of color in different regions were used to calculate this feature parameters which could describe the junction of health and disease. Thirdly, for the design of diseases identification process, the process of diseases was designed based on the key characters of each disease, in this process, diseases were distinguished by the biggest character difference. The concrete step of identification process are as follows: Firstly, color feature was used to divide all the diseases into six groups. Secondly, shape feature was used to subdivide. Finally, texture and junction feature were used for ultimate recognition. Color, shape, texture and junction feature parameters were obtained by related equation, and the parameters were divided into fifteen groups by their disease types. Fourthly, for the establishment of recognition model, the Support Vector Machine(SVM) model was applied to classify and recognize the 15 kinds of rice diseases stepwise. The images were divided into two groups, one was selected to build the model, and another to verify it. LibSVM software package was used to modeling, of which svmtrain function, svmpredict function and grid software were separately used to establish and verify the model and parameter optimization. In this research, a comprehensive study was conducted about the impact of junction feature parameters and identification process on improving accuracy.【Result】A total of 15 kinds of rice disease could be extracted and recognized by this method successfully, and the average recognition accuracy rate was up to 92.67% , average missing report rate was 7.00%, the max missing report rate and false report rate was 15.00% and 25.00%, respectively. After using junction feature parameters, recognition accuracy rate was 14.00% higher, average missing report rate was 7.50% lower, the max decreasing ranges of missing report rate and false report rate were 20.00% and 65.00%, respectively. The identification process proposed in this study could make the recognition accuracy rate 12.67% higher, missing report rate was 9.33% lower, and decreasing ranges of some diseases were more than 30.00%. In all the steps of diseases identification process, the accuracy rate of color recognition was 96.71%, missing report rate and false report rate were both under 10.00%. At the second step, the accuracy rate was 94.17%, missing report rate and false report rate were both under 15.00%. The accuracy rate of texture and junction feature step recognition was 91.5%, missing report rate and false report rate were both under 25.00%.【Conclusion】The 15 kinds of rice disease spots could be segmented from the original images based on the modified mean shift image segmentation algorithm, and the disease could be accurately classified with the method of SVM model with proper feature parameters. The junction feature parameter and the classifying process which were put forward in this paper could improve recognition accuracy. The junction feature parameter could increase the recognition accuracy significantly, especially when the target diseases are similar. Common rice leaf diseases could be recognized by using all those technologies. This paper offered a technical support for automatic diagnoses of rice diseases further.

Key words: rice disease , computer vision , image processing , junction of health and disease , mean shift , support vector machine

刘涛1, 仲晓春2, 孙成明1, 郭文善1, 陈瑛瑛1, 孙娟1. 基于计算机视觉的水稻叶部病害识别研究[J]. 中国农业科学, 2014, 47(4): 664-674.

LIU Tao-1, ZHONG Xiao-Chun-2, SUN Cheng-Ming-1, GUO Wen-Shan-1, CHEN Ying-Ying-1, SUN Juan-1. Recognition of Rice Leaf Diseases Based on Computer Vision[J]. Scientia Agricultura Sinica, 2014, 47(4): 664-674.

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.0578-1752.2014.04.006

https://www.chinaagrisci.com/CN/Y2014/V47/I4/664

参考文献

[1]章毓晋. 时空行为理解. 中国图像图形学报, 2013, 18(2): 141-151.

Zhang Y J. Understanding spatial-temporal behaviors. Journal of Image and Graphics, 2013, 18(2): 141-151. (in Chinese)

[2]赖君臣, 李少昆, 明博, 王娜, 王克如, 谢瑞芝, 高世菊. 作物病害机器视觉诊断研究进展. 中国农业科学, 2009, 42(4): 1215-1221.

Lai J C, Li S K, Ming B, Wang N, Wang K R, Xie R Z, Gao S J. Advances in research on computer-vision diagnosis of crop diseases. Scientia Agricultura Sinica, 2009, 42(4): 1215-1221. (in Chinese)

[3]Camargo A, Smith J. Image pattern classification for the identification of disease causing agents in plants. Computers and Electronics in Agriculture, 2009, 66(2): 121-125.

[4]Camargo A, Smith J. An image-processing based algorithm to automatically identify plant disease visual symptoms. Biosystems Engineering, 2009, 102(1): 9-21.

[5]Pydipati R, Burks T, Lee W. Identification of citrus disease using color texture features and discriminant analysis. Computers and Electronics in Agriculture, 2006, 52(1/2):49-59.

[6]Arribas J, Gonzalo V, Gonzalo R, Jaime G. Leaf classification in sunflower crops by computer vision and neural networks. Computer and Electronics in Agriculture, 2011, 78(1): 9-18.

[7]Wang J, He J, Han Y, Ouyang C, Li D. An Adaptive Thresholding algorithm of field leaf image. Computer and Electronics in Agriculture, 2013, 96: 23-29.

[8]Meunkaewjinda A, Kumsawat P, Attakitmongcol K, Srikaew A. Grape leaf disease detection from color imagery using hybrid intelligent system[C]// Proceedings of ECTl-CON. Krabi: 2008: 513-516.

[9]贾建楠, 吉海彦. 基于病斑形状和神经网络的黄瓜病害识别. 农业工程学报, 2013, 29(增): 115-121.

Jia J N, Ji H Y. Recognition for cucumber disease based on leaf spot shape and neural network. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(S): 115-121. (in Chinese)

[10]陈兵旗, 郭学梅, 李晓华. 基于图像处理的小麦病害诊断算法. 农业机械学报, 2009, 42(12): 190-195.

Chen B Q, Guo X M, Li X H. Image diagnosis algorithm of diseased wheat. Transactions of the Chinese Society for Agricultural Machinery, 2009, 42(12): 190-195. (in Chinese)

[11]温长吉, 王生生, 于合龙, 苏恒强. 基于改进蜂群算法优化神经网络的玉米病害图像分割. 农业工程学报, 2013, 29(13): 142-149.

Wen C J, Wang S S, Yu H L, Su H Q. Image segmentation method for maize diseases based on pulse coupled neural network with modified artificial bee algorithm. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(13): 142-149. (in Chinese)

[12]周正, 雷梦龙, 唐少先. 基于计算机视觉技术的番茄叶部病害识别研究. 安徽农业科学, 2013, 41(5): 2288-2289.

Zhou Z, Lei M L, Tang S X. Study of tomato leaf disease recognition based on computer vision technology. Journal of Anhui Agricultural Sciences, 2013, 41(5): 2288-2289. (in Chinese)

[13]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)

[14]张建华, 祁力均, 冀荣华, 王虎, 黄士凯, 王沛. 基于粗糙集和BP神经网络的棉花病害识别. 农业工程学报, 2012, 28(7): 161-167.

Zhang J H, Qi L J, Ji R H, Wang H, Huang S K, Wang P. Cotton diseases identification based on rough sets and BP neural network. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(7): 161-167. (in Chinese)

[15]石凤梅, 赵开才, 孟庆林, 马立功. 基于支持向量机的水稻稻瘟病图像分割研究. 东北农业大学学报, 2013, 44(2): 128-135.

Shi F M, Zhao K C, Meng Q L, Ma L G. Study on image segmentation of rice blast based on support vector machines. Journal of Northeast Agricultural University, 2013, 44(2): 128-135. (in Chinese)

[16]管泽鑫, 唐健, 杨保军, 周营烽, 范德耀, 姚青. 基于图像的水稻病害识别方法研究. 中国水稻科学, 2010, 24(5): 497-502.

Guan Z X, Tang J, Yang B J, Zhou Y F, Fan D Y, Yao Q. Study on recognition method of rice disease based on image. Chinese Journal of Rice Science, 2010, 24(5): 497-502. (in Chinese)

[17]杨昕薇, 谭峰. 基于贝叶斯分类器的水稻病害识别处理的研究. 黑龙江八一农垦大学学报, 2012, 24(3): 64-67.

Yang X W, Tan F. Research of treatment of the rice disease recognition based on bayes classifier. Journal of Heilongjiang Bayi Agricultural University, 2012, 24(3): 64-67. (in Chinese)

[18]傅强, 黄世文, 谢茂成. 水稻病虫害及防治原色图册. 北京: 金盾出版社, 2009.

Fu Q, Huang S W, Xie M C. Rice Diseases and Insect Pests Atlas. Beijing: Jindun Press, 2009. (in Chinese)

[19]陈福如. 水稻病虫害原色图谱及其诊治技术. 北京: 中国农业科学技术出版社, 2012.

Chen F R. Rice Diseases and Insect Pests Atlas and Control Technology. Beijing: China Agriculture Science Press, 2012. (in Chinese)

[20]夏声广, 唐启义. 水稻病虫草害防治原色生态图谱. 北京: 中国农业出版社, 2006.

Xia S G, Tang Q Y. Rice Disease Pest and Weed Prevention Atlas. Beijing: China Agriculture Press, 2006. (in Chinese)

[21]张铮, 王艳萍, 薛桂香. 数字图像处理与机器视觉Visual C++与Matlab实现. 北京: 人民邮电出版社, 2012.

Zhang Z, Wang Y P, Xue G X. Digital Image Processing and Machine Vision (Visual C++ and Matlab). Beijing: Posts&Telecom Press, 2012. (in Chinese)

[22]Song N, Gu I Y H, Cao Z, Viberg M. Enhance spatial-range mean shift color image segmentation by using convergence frequency and position[C]//Proceeding of the 14th European Signal Processing Conferences, 2006.

[23]隆旺夫. 水稻霜霉病与苗疫霉病的识别和防治技术. 植保技术与推广, 2003, 23(4): 34-35.

Long W F. Rice seedling blight and downy mildew identification and control techniques. Plant Protection Technology and Extension, 2003, 23(4): 34-35. (in Chinese)

[24]王心猛, 王天喜, 武明祥. 水稻霜霉病与赤枯病的识别及防治措施. 植物医生, 2011, 24(2): 4-5.

Wang X M, Wang T X, Wu M X. Downy mildew and red rice blight identification and prevention measures. Plant Doctor, 2011, 24(2): 4-5. (in Chinese)

[25]李俊平, 朱文华. 水稻细菌性褐条病的发生及防治. 贵州农业科学, 2003, 31(2): 34-36.

Li J P, Zhu W H. Occurrence and control of bacterial brown streak in rice. Guizhou Agricultural Sciences, 2003, 31(2): 34-36. (in Chinese)

[26]赖传雅, 袁高庆. 农业植物病理学(华南本). 北京: 科学出版社, 2008.

Lai C Y, Yuan G Q. Agricultural Phytopathology(Southern). Beijing: Science Press, 2008. (in Chinese)

[27]黄志开. 彩色图像特征提取与植物分类研究[D]. 合肥: 中国科学技术大学, 2006.

Huang Z K. A study of feature extraction technique based on color image and plant classification[D]. Hefei: University of Science and Technology of China, 2006. (in Chinese)

[28]汤井田, 胡丹, 龚智敏. 基于SVM的图像纹理特征分类研究. 计算机工程与科学, 2008, 30(8): 44-48.

Tang J T, Hu D, Gong Z M. Research on the classification of SVM-based image texture features. Computer Engineering & Science, 2008, 30(8): 44-48. (in Chinese)

[29]李智峰, 朱谷昌, 董泰锋. 基于灰度共生矩阵的图像纹理特征地物分类应用. 地质与勘探, 2011, 47(3): 456-461.

Li Z F, Zhu G C, Dong T F. Application of GLCM-based texture features to remote sensing image classification. Geology and Exploration, 2011, 47(3): 456-461. (in Chinese)

[30]胡杏春, 徐张芹, 陈杏山, 周永寿. 水稻白叶枯病和稻叶尖枯病的区别. 安徽农业, 1997(7): 20.

Hu X C, Xu Z Q, Chen X S, Zhou Y S. Distinction of rice bacterial leaf blight and rice leaf tip blight. Anhui Agriculture, 1997(7): 20. (in Chinese)

[31]刘占宇, 黄敬峰, 陶荣祥, 张红志. 基于主成分分析和径向基网络的水稻胡麻斑病严重度估测. 光谱学与光谱分析, 2008, 28(9): 2156-2160.

Liu Z Y, Huang J F, Tao R X, Zhang H Z. Estimating rice brown spot disease severity based on principal component analysis and radial basis function neural network. Spectroscopy and Spectral Analysis, 2008, 28(9): 2156-2160. (in Chinese)

[32]张学工. 统计学习理论. 北京: 电子工业出版社, 2004.

Zhang X G. Statistical Learning Theory. Beijing: Publishing House of Electronics Industry, 2004. (in Chinese)

[33]潘明清. 基于支持向量机的机械故障模式分类研究[D]. 杭州: 浙江大学, 2005.

Pan M Q. Research on machine fault pattern classification based on support vector machine[D]. Hangzhou: Zhejiang University, 2005. (in Chinese)