基于椭圆型度量学习的小麦叶部病害识别
|
摘要
特征提取和相似性度量是基于图像处理的农作物病虫害识别方法中的两大关键问题。以感染小麦白粉病的叶片为研究对象,提出了一种基于椭圆型度量学习的小麦叶部病害严重度识别算法。该算法首先给出了一种滑窗最大值(Moving window maximum,MWM)特征提取方法,对分割后的病斑图像采用滑窗法提取HSV颜色特征和LBP纹理特征,在同一水平条滑窗上取每一维特征的最大值作为这一水平条的特征,这种MWM特征表示方法能有效减弱小麦叶片弯曲、倾斜、拍摄角度不同等对识别率的影响;然后,引入对样本数据具有更好区分性的椭圆型度量,根据样本的类内与类间高斯分布的对数似然比定义椭圆型度量矩阵,为了保持最大化的分类信息,将特征子空间学习和椭圆型度量学习同时进行;最后,利用得到的椭圆型度量计算特征向量之间的距离实现不同严重度病害的识别。对比实验结果表明,本文算法使得小麦白粉病严重度的识别正确率达到了100%,优于SVM方法的88. 33%、BP神经网络方法的90%。
Feature extraction and similarity measurement are two key problems of crop pest recognition based on image processing. The leaves of wheat powdery mildew were treated as the research objects,and an algorithm of wheat leaf disease severity recognition based on elliptical metric learning was proposed.Firstly,a method of moving window maximum(MWM) feature extraction was presented in the algorithm.The HSV color features and LBP texture features were extracted by using the sliding window method from the segmented lesion images. The maximum value of each dimension feature on the same horizontal sliding window was taken as the feature of this horizontal bar. The MWM feature representation method can effectively reduce the influence of curvature,tilt and different shooting angles of wheat leaves on the recognition rate. Then,an elliptical metric with better distinguishability for sample data was introduced,and the elliptic metric matrix was defined based on the log-likelihood ratio of Gaussian distributions on the intrapersonal sample and the extrapersonal sample. In order to maintain the maximal classification information,the feature subspace learning and elliptic metric learning were performed simultaneously.Finally,to recognize the severity of diseases,the elliptic metric was used to calculate the distance between the eigenvectors. The results of comparison experiments showed that the recognition rate of wheat powdery mildew severity was 100%,which was better than 88. 33% for SVM method and 90% for BP neural network method. The research result can provide valuable help for the intelligent recognition of crop disease severity.
Feature extraction and similarity measurement are two key problems of crop pest recognition based on image processing. The leaves of wheat powdery mildew were treated as the research objects,and an algorithm of wheat leaf disease severity recognition based on elliptical metric learning was proposed.Firstly,a method of moving window maximum(MWM) feature extraction was presented in the algorithm.The HSV color features and LBP texture features were extracted by using the sliding window method from the segmented lesion images. The maximum value of each dimension feature on the same horizontal sliding window was taken as the feature of this horizontal bar. The MWM feature representation method can effectively reduce the influence of curvature,tilt and different shooting angles of wheat leaves on the recognition rate. Then,an elliptical metric with better distinguishability for sample data was introduced,and the elliptic metric matrix was defined based on the log-likelihood ratio of Gaussian distributions on the intrapersonal sample and the extrapersonal sample. In order to maintain the maximal classification information,the feature subspace learning and elliptic metric learning were performed simultaneously.Finally,to recognize the severity of diseases,the elliptic metric was used to calculate the distance between the eigenvectors. The results of comparison experiments showed that the recognition rate of wheat powdery mildew severity was 100%,which was better than 88. 33% for SVM method and 90% for BP neural network method. The research result can provide valuable help for the intelligent recognition of crop disease severity.
引文
1王翔宇,温皓杰,李鑫星,等.农业主要病害检测与预警技术研究进展分析[J/OL].农业机械学报,2016,47(9):266-277.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20160937&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2016.09.037.WANG Xiangyu,WEN Haojie,LI Xinxing,et al.Analysis of the research progress on the detection and early warning technology of major agricultural diseases[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2016,47(9):266-277.(in Chinese)
2 ZHANG S W,WANG Z,WANG X F,et al.Crop disease recognition based on leaf image processing[C]∥Proceedings of 2014International Conference on Environmental Science and Biological Engineering,2015:889-896.
3张竞成,袁琳,王纪华,等.作物病虫害遥感监测研究进展[J].农业工程学报,2012,28(20):1-11.ZHANG Jingcheng,YUAN Lin,WANG Jihua,et al.Advances in remote sensing monitoring of crop diseases and insect pests[J].Transactions of the CSAE,2012,28(20):1-11.(in Chinese)
4 RONIS A,SEMAKIENE R,DABKEVIˇCIUS Z,et al.Influence of leaf diseases on grain yield and yield components in winter wheat[J].Journal of Plant Protection Research,2009,49(2):151-157.
5 SETHY P K,NEGI B,BARPANDA N K,et al.Measurement of disease severity of rice crop using machine learning and computational intelligence[M].Cognitive Science and Artificial Intelligence,Springer,2018:1-11.
6肖志云,刘洪.马铃薯典型病害图像自适应特征融合与快速识别[J/OL].农业机械学报,2017,48(12):26-32.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20171203&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2017.12.003.XIAO Zhiyun,LIU Hong.Adaptive feature fusion and rapid identification of potato typical diseases image[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2017,48(12):26-32.(in Chinese)
7任守纲,陆海飞,袁培森,等.基于显著性检测的黄瓜叶部病害图像分割算法[J/OL].农业机械学报,2016,47(9):11-16.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20160902&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2016.09.002.REN Shougang,LU Haifei,YUAN Peisen,et al.Image segmentation algorithm for cucumber leaf diseases based on saliency detection[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2016,47(9):11-16.(in Chinese)
8田有文,李天来,李成华,等.基于支持向量机的葡萄病害图像识别方法[J].农业工程学报,2007,23(6):175-180.TIAN Youwen,LI Tianlai,LI Chenghua,et al.Image recognition method of grape diseases based on support vector machine[J].Transactions of the CSAE,2007,23(6):175-180.(in Chinese)
9赵进辉,罗锡文,周志艳.基于颜色与形状特征的甘蔗病害图像分割方法[J].农业机械学报,2008,39(9):100-103.ZHAO Jinhui,LUO Xiwen,ZHOU Zhiyan.Image segmentation method of sugarcane disease based on color and shape characteristics[J].Transactions of the Chinese Society for Agricultural Machinery,2008,39(9):100-103.(in Chinese)
10 ZHANG S,WANG H,HUANG W,et al.Plant diseased leaf segmentation and recognition by fusion of superpixel,K-means and PHOG[J].Optik-International Journal for Light and Electron Optics,2017,157:866-872.
11邓继忠,李敏,袁之报,等.基于图像识别的小麦腥黑穗病害诊断技术研究[J].东北农业大学学报,2012,43(5):74-77.DENG Jizhong,LI Min,YUAN Zhibao,et al.Research on diagnosis of wheat smut disease based on image recognition[J].Journal of Northeast Agricultural University,2012,43(5):74-77.(in Chinese)
12田杰.基于图像分析的小麦叶部病害识别方法研究[D].杨凌:西北农林科技大学,2015.TIAN Jie.Wheat leaf disease identification method based on image analysis[D].Yangling:Northwest A&F University,2015.(in Chinese)
13 PHADIKAR S,SIL J.Rice disease identification using pattern recognition techniques[C]∥International Conference on Computer and Information Technology.IEEE,2008:420-423.
14许良凤,徐小兵,胡敏,等.基于多分类器融合的玉米叶部病害识别[J].农业工程学报,2015,31(14):194-201,315.XU Liangfeng,XU Xiaobing,HU Min,et al.Corn leaf diseases identification based on multiple classifiers fusion[J].Transactions of the CSAE,2015,31(14):194-201,315.(in Chinese)
15黄双萍,齐龙,马旭,等.基于高光谱成像的水稻穗瘟病害程度分级方法[J].农业工程学报,2015,31(1):212-219.HUANG Shuangping,QI Long,MA Xu,et al.Classification of rice panicle blast diseases based on hyperspectral imaging[J].Transactions of the CSAE,2015,31(1):212-219.(in Chinese)
16张善文,张传雷.基于局部判别映射算法的玉米病害识别方法[J].农业工程学报,2014,30(11):167-172.ZHANG Shanwen,ZHANG Chuanlei.Maize disease recognition method based on local discriminant mapping algorithm[J].Transactions of the CSAE,2014,30(11):167-172.(in Chinese)
17王献锋,张善文,王震,等.基于叶片图像和环境信息的黄瓜病害识别方法[J].农业工程学报,2014,30(14):148-153.WANG Xianfeng,ZHANG Shanwen,WANG Zhen,et al.Cucumber disease identification method based on leaf image and environmental information[J].Transactions of the CSAE,2014,30(14):148-153.(in Chinese)
18张建华,孔繁涛,李哲敏,等..基于最优二叉树支持向量机的蜜柚叶部病害识别[J].农业工程学报,2014,30(19):222-231.ZHANG Jianhua,KONG Fantao,LI Zhemin,et al.Identification of pomelo leaf diseases based on the optimal two branch tree support vector[J].Transactions of the CSAE,2014,30(19):222-231.(in Chinese)
19 HAN L,HALEEM M S,TAYLOR M.A novel computer vision-based approach to automatic detection and severity assessment of crop diseases[C]∥Science and Information Conference.IEEE,2015:638-644.
20 PYDIPATI R,BURKS T F,LEE W S.Identification of citrus disease using color texture features and discriminant analysis[J].Computers and Electronics in Agriculture,2006,52(1):49-59.
21 ZHANG Dongyan,XIE Xinhua,LIANG Dong,et al.Comparative analysis of identification methods for wheat leaf diseases[J].International Agricultural Engineering Journal,2016,25(3):197-203.
22王卫东,杨静宇.采用虚拟训练样本的二次判别分析方法[J].自动化学报,2008,34(4):400-407.WANG Weidong,YANG Jingyu.Two discriminant analysis method using virtual training samples[J].Acta Automatica Sinica,2008,34(4):400-407.(in Chinese)
23蒋龙泉,鲁帅,冯瑞,等.基于多特征融合和SVM分类器的植物病虫害检测方法[J].计算机应用与软件,2014,31(12):186-190.JIANG Longquan,LU Shuai,FENG Rui,et al.Plant pest and disease detection method based on multi feature fusion and SVMclassifier[J].Computer Applications and Software,2014,31(12):186-190.(in Chinese)
24张永玲.基于Android的水稻病虫害图像识别与诊断系统的研究[D].杭州:浙江理工大学,2018.ZHANG Yongling.Research on image recognition and diagnosis system of rice diseases and insect pests based on Android[D].Hangzhou:Zhejiang Sci-Tech University,2018.(in Chinese)
2 ZHANG S W,WANG Z,WANG X F,et al.Crop disease recognition based on leaf image processing[C]∥Proceedings of 2014International Conference on Environmental Science and Biological Engineering,2015:889-896.
3张竞成,袁琳,王纪华,等.作物病虫害遥感监测研究进展[J].农业工程学报,2012,28(20):1-11.ZHANG Jingcheng,YUAN Lin,WANG Jihua,et al.Advances in remote sensing monitoring of crop diseases and insect pests[J].Transactions of the CSAE,2012,28(20):1-11.(in Chinese)
4 RONIS A,SEMAKIENE R,DABKEVIˇCIUS Z,et al.Influence of leaf diseases on grain yield and yield components in winter wheat[J].Journal of Plant Protection Research,2009,49(2):151-157.
5 SETHY P K,NEGI B,BARPANDA N K,et al.Measurement of disease severity of rice crop using machine learning and computational intelligence[M].Cognitive Science and Artificial Intelligence,Springer,2018:1-11.
6肖志云,刘洪.马铃薯典型病害图像自适应特征融合与快速识别[J/OL].农业机械学报,2017,48(12):26-32.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20171203&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2017.12.003.XIAO Zhiyun,LIU Hong.Adaptive feature fusion and rapid identification of potato typical diseases image[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2017,48(12):26-32.(in Chinese)
7任守纲,陆海飞,袁培森,等.基于显著性检测的黄瓜叶部病害图像分割算法[J/OL].农业机械学报,2016,47(9):11-16.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?flag=1&file_no=20160902&journal_id=jcsam.DOI:10.6041/j.issn.1000-1298.2016.09.002.REN Shougang,LU Haifei,YUAN Peisen,et al.Image segmentation algorithm for cucumber leaf diseases based on saliency detection[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2016,47(9):11-16.(in Chinese)
8田有文,李天来,李成华,等.基于支持向量机的葡萄病害图像识别方法[J].农业工程学报,2007,23(6):175-180.TIAN Youwen,LI Tianlai,LI Chenghua,et al.Image recognition method of grape diseases based on support vector machine[J].Transactions of the CSAE,2007,23(6):175-180.(in Chinese)
9赵进辉,罗锡文,周志艳.基于颜色与形状特征的甘蔗病害图像分割方法[J].农业机械学报,2008,39(9):100-103.ZHAO Jinhui,LUO Xiwen,ZHOU Zhiyan.Image segmentation method of sugarcane disease based on color and shape characteristics[J].Transactions of the Chinese Society for Agricultural Machinery,2008,39(9):100-103.(in Chinese)
10 ZHANG S,WANG H,HUANG W,et al.Plant diseased leaf segmentation and recognition by fusion of superpixel,K-means and PHOG[J].Optik-International Journal for Light and Electron Optics,2017,157:866-872.
11邓继忠,李敏,袁之报,等.基于图像识别的小麦腥黑穗病害诊断技术研究[J].东北农业大学学报,2012,43(5):74-77.DENG Jizhong,LI Min,YUAN Zhibao,et al.Research on diagnosis of wheat smut disease based on image recognition[J].Journal of Northeast Agricultural University,2012,43(5):74-77.(in Chinese)
12田杰.基于图像分析的小麦叶部病害识别方法研究[D].杨凌:西北农林科技大学,2015.TIAN Jie.Wheat leaf disease identification method based on image analysis[D].Yangling:Northwest A&F University,2015.(in Chinese)
13 PHADIKAR S,SIL J.Rice disease identification using pattern recognition techniques[C]∥International Conference on Computer and Information Technology.IEEE,2008:420-423.
14许良凤,徐小兵,胡敏,等.基于多分类器融合的玉米叶部病害识别[J].农业工程学报,2015,31(14):194-201,315.XU Liangfeng,XU Xiaobing,HU Min,et al.Corn leaf diseases identification based on multiple classifiers fusion[J].Transactions of the CSAE,2015,31(14):194-201,315.(in Chinese)
15黄双萍,齐龙,马旭,等.基于高光谱成像的水稻穗瘟病害程度分级方法[J].农业工程学报,2015,31(1):212-219.HUANG Shuangping,QI Long,MA Xu,et al.Classification of rice panicle blast diseases based on hyperspectral imaging[J].Transactions of the CSAE,2015,31(1):212-219.(in Chinese)
16张善文,张传雷.基于局部判别映射算法的玉米病害识别方法[J].农业工程学报,2014,30(11):167-172.ZHANG Shanwen,ZHANG Chuanlei.Maize disease recognition method based on local discriminant mapping algorithm[J].Transactions of the CSAE,2014,30(11):167-172.(in Chinese)
17王献锋,张善文,王震,等.基于叶片图像和环境信息的黄瓜病害识别方法[J].农业工程学报,2014,30(14):148-153.WANG Xianfeng,ZHANG Shanwen,WANG Zhen,et al.Cucumber disease identification method based on leaf image and environmental information[J].Transactions of the CSAE,2014,30(14):148-153.(in Chinese)
18张建华,孔繁涛,李哲敏,等..基于最优二叉树支持向量机的蜜柚叶部病害识别[J].农业工程学报,2014,30(19):222-231.ZHANG Jianhua,KONG Fantao,LI Zhemin,et al.Identification of pomelo leaf diseases based on the optimal two branch tree support vector[J].Transactions of the CSAE,2014,30(19):222-231.(in Chinese)
19 HAN L,HALEEM M S,TAYLOR M.A novel computer vision-based approach to automatic detection and severity assessment of crop diseases[C]∥Science and Information Conference.IEEE,2015:638-644.
20 PYDIPATI R,BURKS T F,LEE W S.Identification of citrus disease using color texture features and discriminant analysis[J].Computers and Electronics in Agriculture,2006,52(1):49-59.
21 ZHANG Dongyan,XIE Xinhua,LIANG Dong,et al.Comparative analysis of identification methods for wheat leaf diseases[J].International Agricultural Engineering Journal,2016,25(3):197-203.
22王卫东,杨静宇.采用虚拟训练样本的二次判别分析方法[J].自动化学报,2008,34(4):400-407.WANG Weidong,YANG Jingyu.Two discriminant analysis method using virtual training samples[J].Acta Automatica Sinica,2008,34(4):400-407.(in Chinese)
23蒋龙泉,鲁帅,冯瑞,等.基于多特征融合和SVM分类器的植物病虫害检测方法[J].计算机应用与软件,2014,31(12):186-190.JIANG Longquan,LU Shuai,FENG Rui,et al.Plant pest and disease detection method based on multi feature fusion and SVMclassifier[J].Computer Applications and Software,2014,31(12):186-190.(in Chinese)
24张永玲.基于Android的水稻病虫害图像识别与诊断系统的研究[D].杭州:浙江理工大学,2018.ZHANG Yongling.Research on image recognition and diagnosis system of rice diseases and insect pests based on Android[D].Hangzhou:Zhejiang Sci-Tech University,2018.(in Chinese)