改进自适应分水岭方法分割棉花叶部粘连病斑
|
摘要
针对棉花叶部病斑相互之间存在粘连问题,该文提出了一种自适应分水岭分割方法。该方法在H-minima分水岭分割方法基础上,结合最小二乘圆法误差理论,对图像中每个连通分量进行最小二乘圆拟合,并计算最小二乘圆误差值,通过最小二乘圆误差值大小判断每个连通分量的轮廓不规则度,针对不同轮廓不规则度确定H-minima变换的极小值阈值,根据不同极小值阈值实现棉花叶部粘连病斑的分水岭分割。不同数量粘连病斑分割试验结果表明:该方法实现了棉花叶部粘连病斑数量从2个粘连至5个粘连病斑的自动分割,分割准确率为91.25%,平均运行时间为0.088 s。不同分割方法对比结果显示:该方法能实现对棉花轮纹病、褐斑病、炭疽病、叶斑病和棉铃疫病共5种病害的粘连病斑自动分割,并将距离分水岭分割方法、梯度分水岭分割方法、标记分水岭分割方法、Chan-Vese方法、高斯混合方法与该文方法比较,正确分割率分别为67.8%、36.4%、83.7%、70.3%、82.1%、93.5%,该方法优于其他5种分割方法,有效抑制了过分割问题;在复杂背景、光照不均匀、病斑大小不一致等复杂条件下,该文方法也能较好地实现粘连病斑的分割。该方法不仅能对棉花叶部粘连病斑自动分割,也能为其他作物叶片粘连病斑分割提供参考。
Accurate segmentation of cotton leaf adhesion lesions not only can extract the feature vector of individual lesions to a large extent, but also is important for the improvement of the accuracy of lesion identification. Some methods for adhesion lesions segmentation may result in over-segmentation or under-segmentation. In addition, some segmentation methods for adhesion lesions can better solve the segmentation problem, but the selection of parameters is sensitive, which needs to be manually set and cannot adapt to complex conditions, such as H-minima transform. Aiming at the adhesion problem between lesions, an adaptive watershed segmentation method was proposed. Firstly, for the cotton lesion area extraction, the Gaussian filter was used for image filtering processing, and then the super green color component was extracted and OTSU binary segmentation was performed. Post-segmentation processing was carried by the mathematical morphology of hole filling and the morphological opening operation, so that the contour of segmented lesion area was continuous and the edge was smooth. Secondly, the local minimum threshold(h) was determined. Based on the H-minima watershed segmentation method, the proposed method combined the least squares method error theory to fit the least squares of each connected component in the image. And the least squares error value was calculated, then the contour irregularity of each connected component was determined according to the least squares error value. The minimum threshold h of the H-minima transform was determined based on different contour irregularities. Finally, Watershed segmentation of cotton leaf adhesion lesions was achieved based on different minimum thresholds. A total of 160 images with 2 to 5 adhesion lesions were selected from the lesion sample library for testing. The lesions segmentation accuracy for cotton leaves with different number of adhesion lesions was 91.25% with running time of 0.088 s. The proposal method achieved the automatic segmentation, and was especially suitable for the adhesion of different lesions and irregularities. Meanwhile, 150 samples with high degree adhesion lesion of ring disease, brown spot, anthracnose, leaf spot and cotton boll blight were selected for different segmentation methods contrast test. The results of contrast test-showed that the proposed method could automatically segment the adhesion lesions of 5 diseases. Distance watershed method, Gradient watershed segmentation method, marker watershed segmentation method, Chan-Vese method, Gaussian mixture method and the proposed method, the correct segmentation ratio were 67.8%, 36.4%, 83.7%, 70.3%, 82.1% and 93.5%, respectively, with the average running time of 0.034, 0.036, 0.046, 0.357, 0.108 and 0.094 s. The experimental results showed that the proposed method was superior to the other 5 methods, especially the over-segmentation problem was effectively suppressed. The proposed method took slightly longer time than the distance watershed segmentation method, the gradient watershed segmentation method, and the marker watershed segmentation method, and was lower than the Chan-Vese method and the Gaussian mixture method, which could still meet the real-time image processing requirements. The results of lesion segmentation test in complex environment showed that under complex conditions such as complex background, uneven illumination and uniform lesion size, the proposed method could better achieve the segmentation of adhesion lesions, and its segmentation accuracy and running time could meet the actual needs. The proposed method can not only automatically segment the adhesion lesions on cotton leaf, but also provide reference for the segmentation of adhesion lesions for other crop leaves.
Accurate segmentation of cotton leaf adhesion lesions not only can extract the feature vector of individual lesions to a large extent, but also is important for the improvement of the accuracy of lesion identification. Some methods for adhesion lesions segmentation may result in over-segmentation or under-segmentation. In addition, some segmentation methods for adhesion lesions can better solve the segmentation problem, but the selection of parameters is sensitive, which needs to be manually set and cannot adapt to complex conditions, such as H-minima transform. Aiming at the adhesion problem between lesions, an adaptive watershed segmentation method was proposed. Firstly, for the cotton lesion area extraction, the Gaussian filter was used for image filtering processing, and then the super green color component was extracted and OTSU binary segmentation was performed. Post-segmentation processing was carried by the mathematical morphology of hole filling and the morphological opening operation, so that the contour of segmented lesion area was continuous and the edge was smooth. Secondly, the local minimum threshold(h) was determined. Based on the H-minima watershed segmentation method, the proposed method combined the least squares method error theory to fit the least squares of each connected component in the image. And the least squares error value was calculated, then the contour irregularity of each connected component was determined according to the least squares error value. The minimum threshold h of the H-minima transform was determined based on different contour irregularities. Finally, Watershed segmentation of cotton leaf adhesion lesions was achieved based on different minimum thresholds. A total of 160 images with 2 to 5 adhesion lesions were selected from the lesion sample library for testing. The lesions segmentation accuracy for cotton leaves with different number of adhesion lesions was 91.25% with running time of 0.088 s. The proposal method achieved the automatic segmentation, and was especially suitable for the adhesion of different lesions and irregularities. Meanwhile, 150 samples with high degree adhesion lesion of ring disease, brown spot, anthracnose, leaf spot and cotton boll blight were selected for different segmentation methods contrast test. The results of contrast test-showed that the proposed method could automatically segment the adhesion lesions of 5 diseases. Distance watershed method, Gradient watershed segmentation method, marker watershed segmentation method, Chan-Vese method, Gaussian mixture method and the proposed method, the correct segmentation ratio were 67.8%, 36.4%, 83.7%, 70.3%, 82.1% and 93.5%, respectively, with the average running time of 0.034, 0.036, 0.046, 0.357, 0.108 and 0.094 s. The experimental results showed that the proposed method was superior to the other 5 methods, especially the over-segmentation problem was effectively suppressed. The proposed method took slightly longer time than the distance watershed segmentation method, the gradient watershed segmentation method, and the marker watershed segmentation method, and was lower than the Chan-Vese method and the Gaussian mixture method, which could still meet the real-time image processing requirements. The results of lesion segmentation test in complex environment showed that under complex conditions such as complex background, uneven illumination and uniform lesion size, the proposed method could better achieve the segmentation of adhesion lesions, and its segmentation accuracy and running time could meet the actual needs. The proposed method can not only automatically segment the adhesion lesions on cotton leaf, but also provide reference for the segmentation of adhesion lesions for other crop leaves.
引文
[1]Lu J,Hu J,Zhao G,et al.An in-field automatic wheat disease diagnosis system[J].Computers&Electronics in Agriculture,2017,142(9):369-379.
[2]Zhang J,Kong F,Zhai Z,et al.Robust image segmentation method for cotton leaf under natural conditions based on immune algorithm and PCNN algorithm[J].International Journal of Pattern Recognition&Artificial Intelligence,2018,32(5):1-22.
[3]Barbedo J G A.A review on the main challenges in automatic plant disease identification based on visible range images[J].Biosystems Engineering,2016,144(1):52-60.
[4]Bock C H,Poole G H,Parker P E,et al.Plant disease severity estimated visually,by digital photography and image analysis,and by hyperspectral imaging[J].Critical Reviews in Plant Sciences,2010,29(2):59-107.
[5]Zhang J,Kong F,Zhai Z,et al.Automatic image segmentation method for cotton leaves with disease under natural environment[J].Journal of Integrative Agriculture2018,17(8):1800-1814.
[6]Phadikar S,Sil J,Das A K.Rice diseases classification using feature selection and rule generation techniques[J].Computers&Electronics in Agriculture,2013,90(3):76-85.
[7]张艳诚,毛罕平,胡波,等.作物病害图像中重叠病斑分离算法[J].农业机械学报,2008,39(2):112-115.Zhang Yancheng,Mao Hanping,Hu Bo,et al.Separate algorithm for overlapping spots in crop disease image[J].Transactions of the Chinese Society for Agricultural Machinery,2008,39(2):112-115.(in Chinese with English abstract)
[8]邱白晶,王天波,李娟娟,等.黄瓜蚜虫的图像识别与计数方法[J].农业机械学报,2010,41(8):151-155.Qiu Baijing,Wang Tianbo,Li Juanjuan,et al.Image recognition and counting for glasshouse Aphis gossypii[J].Transactions of the Chinese Society for Agricultural Machinery,2010,41(8):151-155.(in Chinese with English abstract)
[9]任艳娜,席磊,汪强,等.基于高斯概率模型的水稻重叠病斑分割算法[J].计算机仿真,2011,28(2):341-344.Ren Yanna,Xi Lei,Wang Qiang,et al.Design of rice lesion identification system based embedded[J].Computer Simulation,2011,28(2):341-344.(in Chinese with English abstract)
[10]张晴晴,齐国红,张云龙.基于改进分水岭算法的作物病害叶片分割方法[J].江苏农业科学,2015,43(2):400-403.Zhang Qingqing,Qi Guohong,Zhang Yunlong,et al.Crop disease leaf segmentation method based on improved watershed algorithm[J].Jiangsu Agricultural Science,2015,43(2):400-403.(in Chinese with English abstract)
[11]米雅婷.基于GA-BP神经网络的温室番茄病害诊断研究[D].哈尔滨:东北林业大学,2016.Mi Yating.Research on Greenhouse Tomato Disease Diagnosis based on GA-BP Network[D].Harbin:Northeast Forestry University,2016.(in Chinese with English abstract)
[12]Li Y,Xia C,Lee J.Detection of small-sized insect pest in greenhouses based on multifractal analysis[J].OptikInternational Journal for Light and Electron Optics,2015,126(19):2138-2143.
[13]Ding W,Taylor G.Automatic moth detection from trap images for pest management[J].Computers&Electronics in Agriculture,2016,123(3):17-28.
[14]Yao Q,Liu Q,Dietterich T G,et al.Segmentation of touching insects based on optical flow and NCuts[J].Biosystems Engineering,2013,114(2):67-77.
[15]Wang Z,Wang K,Yang F,et al.Image segmentation of overlapping leaves based on chan-vese model and sobeloperator[J].Information Processing in Agriculture,2018,5(1):1-10.
[16]Fang Y,Ramasamy R P.Current and prospective methods for plant disease detection[J].Biosensors,2015,5(3):537-561.
[17]Melo G J A D,Gomes V,Baccili C C,et al.A robust segmentation method for counting bovine milk somatic cells in microscope slide images[J].Computers&Electronics in Agriculture,2015,115(5):142-149.
[18]Xia C,Chon T S,Ren Z,et al.Automatic identification and counting of small size pests in greenhouse conditions with low computational cost[J].Ecological Informatics,2015,29(2):139-146.
[19]李文勇,李明,钱建平,等.基于形状因子和分割点定位的粘连害虫图像分割方法[J].农业工程学报,2015,31(5):175-180.Li Wenyong,Li Ming,Qian Jianping,et al.Segmentation method for touching pest images based on shape factor and separation points location[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(5):175-180.(in Chinese with English abstract)
[20]李凯,冯全,张建华.棉花苗叶片复杂背景图像的联合分割算法[J].计算机辅助设计与图形学学报,2017,29(10):1871-1880.Li Kai,Feng Quan,Zhang Jianhua.Co-segmentation algorithm for complex background image of cotton seedling leaves[J].Journal of Computer-Aided Design and Computer Graphics,2017,29(10):1871-1880.(in Chinese with English abstract)
[21]Bhoyar K K,Kakde O G.Color image segmentation based on JND color histogram[J].International Journal of Image Processing,2010,3(6):282-293.
[22]Bhogal A K,Singla N,Kaur M.Color image segmentation based on color and texture properties[J].International Journal of Advanced Engineering Sciences and Technologies,2011,8(2):152-159.
[23]李凯,张建华,冯全,等.复杂背景与天气条件下的棉花叶片图像分割方法[J].中国农业大学学报,2018,23(2):88-98.Li Kai,Zhang Jianhua,Feng Quan,et al.Image segmentation method complex background and for cotton leaf under weather conditions[J].Journal of China Agricultural University,2018,23(2):88-98.(in Chinese with English abstract)
[24]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.Recognition of honey pomelo leaf diseases based on optimal binary tree support vector machine[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2014,30(19):222-231.(in Chinese with English abstract)
[25]江海洋,张建,袁媛,等.基于MDMP-LSM算法的黄瓜叶片病斑分割方法[J].农业工程学报,2012,28(21):142-148.Jiang Haiyang,Zhang Jian,Yuan Yuan,et al.Segmentation of cucumber disease leaf image based on MDMP-LSM[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2012,28(21):142-148.(in Chinese with English abstract)
[26]Ren Y G,Jian Z,Miao L,et al.Segmentation method for crop disease leaf images based on watershed algorithm[J].Journal of Computer Applications,2012,32(3):752-755.
[27]Liu T,Chen W,Wu W,et al.Detection of aphids in wheat fields using a computer vision technique[J].Biosystems Engineering,2016,141(11):82-93.
[28]王娅.血液红细胞图像自适应标记分水岭分割算法[J].中国图象图形学报,2017,22(12):1779-1787.Wang Ya.Adaptive marked watershed segmentation algorithm for red blood cell images[J].Journal of Image&Graphics,2017,22(12):1779-1787.(in Chinese with English abstract)
[29]方红萍,方康玲,刘新海,等.自适应H-minima的改进分水岭堆叠细胞分割方法[J].计算机应用研究,2016,33(5):1587-1590.Fang Hongping,Fang Kangling,Liu Xinhai,et al.Clustered cells segmentation using modified watershed method based on adaptive H-minima[J].Application Research of Computers,2016,33(5):1587-1590.(in Chinese with English abstract)
[30]石际亮,董黎君,梁国星.基于MATLAB图像处理的圆度误差数据采集方法[J].中国农机化学报,2015,36(2):257-261.Shi Jiliang,Dong Lijun,Liang Guoxing.Acquisition method of roundness error data based on MATLAB image processing[J].Journal of Chinese Agricultural Mechanization,2015,36(2):257-261.(in Chinese with English abstract)
[31]岳奎.最小二乘圆法评定圆度误差的程序设计[J].工具技术,2006,40(4):79-81.Yue Kui.Programming for evaluation of roundness error by least squares mean circle method[J].Tool Engineering,2006,40(4):79-81.(in Chinese with English abstract)
[32]Wang J,He J,Han Y,et al.An adaptive thresholding algorithm of field leaf image[J].Computers and Electronics in Agriculture,2013,96(6):23-39.
[2]Zhang J,Kong F,Zhai Z,et al.Robust image segmentation method for cotton leaf under natural conditions based on immune algorithm and PCNN algorithm[J].International Journal of Pattern Recognition&Artificial Intelligence,2018,32(5):1-22.
[3]Barbedo J G A.A review on the main challenges in automatic plant disease identification based on visible range images[J].Biosystems Engineering,2016,144(1):52-60.
[4]Bock C H,Poole G H,Parker P E,et al.Plant disease severity estimated visually,by digital photography and image analysis,and by hyperspectral imaging[J].Critical Reviews in Plant Sciences,2010,29(2):59-107.
[5]Zhang J,Kong F,Zhai Z,et al.Automatic image segmentation method for cotton leaves with disease under natural environment[J].Journal of Integrative Agriculture2018,17(8):1800-1814.
[6]Phadikar S,Sil J,Das A K.Rice diseases classification using feature selection and rule generation techniques[J].Computers&Electronics in Agriculture,2013,90(3):76-85.
[7]张艳诚,毛罕平,胡波,等.作物病害图像中重叠病斑分离算法[J].农业机械学报,2008,39(2):112-115.Zhang Yancheng,Mao Hanping,Hu Bo,et al.Separate algorithm for overlapping spots in crop disease image[J].Transactions of the Chinese Society for Agricultural Machinery,2008,39(2):112-115.(in Chinese with English abstract)
[8]邱白晶,王天波,李娟娟,等.黄瓜蚜虫的图像识别与计数方法[J].农业机械学报,2010,41(8):151-155.Qiu Baijing,Wang Tianbo,Li Juanjuan,et al.Image recognition and counting for glasshouse Aphis gossypii[J].Transactions of the Chinese Society for Agricultural Machinery,2010,41(8):151-155.(in Chinese with English abstract)
[9]任艳娜,席磊,汪强,等.基于高斯概率模型的水稻重叠病斑分割算法[J].计算机仿真,2011,28(2):341-344.Ren Yanna,Xi Lei,Wang Qiang,et al.Design of rice lesion identification system based embedded[J].Computer Simulation,2011,28(2):341-344.(in Chinese with English abstract)
[10]张晴晴,齐国红,张云龙.基于改进分水岭算法的作物病害叶片分割方法[J].江苏农业科学,2015,43(2):400-403.Zhang Qingqing,Qi Guohong,Zhang Yunlong,et al.Crop disease leaf segmentation method based on improved watershed algorithm[J].Jiangsu Agricultural Science,2015,43(2):400-403.(in Chinese with English abstract)
[11]米雅婷.基于GA-BP神经网络的温室番茄病害诊断研究[D].哈尔滨:东北林业大学,2016.Mi Yating.Research on Greenhouse Tomato Disease Diagnosis based on GA-BP Network[D].Harbin:Northeast Forestry University,2016.(in Chinese with English abstract)
[12]Li Y,Xia C,Lee J.Detection of small-sized insect pest in greenhouses based on multifractal analysis[J].OptikInternational Journal for Light and Electron Optics,2015,126(19):2138-2143.
[13]Ding W,Taylor G.Automatic moth detection from trap images for pest management[J].Computers&Electronics in Agriculture,2016,123(3):17-28.
[14]Yao Q,Liu Q,Dietterich T G,et al.Segmentation of touching insects based on optical flow and NCuts[J].Biosystems Engineering,2013,114(2):67-77.
[15]Wang Z,Wang K,Yang F,et al.Image segmentation of overlapping leaves based on chan-vese model and sobeloperator[J].Information Processing in Agriculture,2018,5(1):1-10.
[16]Fang Y,Ramasamy R P.Current and prospective methods for plant disease detection[J].Biosensors,2015,5(3):537-561.
[17]Melo G J A D,Gomes V,Baccili C C,et al.A robust segmentation method for counting bovine milk somatic cells in microscope slide images[J].Computers&Electronics in Agriculture,2015,115(5):142-149.
[18]Xia C,Chon T S,Ren Z,et al.Automatic identification and counting of small size pests in greenhouse conditions with low computational cost[J].Ecological Informatics,2015,29(2):139-146.
[19]李文勇,李明,钱建平,等.基于形状因子和分割点定位的粘连害虫图像分割方法[J].农业工程学报,2015,31(5):175-180.Li Wenyong,Li Ming,Qian Jianping,et al.Segmentation method for touching pest images based on shape factor and separation points location[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(5):175-180.(in Chinese with English abstract)
[20]李凯,冯全,张建华.棉花苗叶片复杂背景图像的联合分割算法[J].计算机辅助设计与图形学学报,2017,29(10):1871-1880.Li Kai,Feng Quan,Zhang Jianhua.Co-segmentation algorithm for complex background image of cotton seedling leaves[J].Journal of Computer-Aided Design and Computer Graphics,2017,29(10):1871-1880.(in Chinese with English abstract)
[21]Bhoyar K K,Kakde O G.Color image segmentation based on JND color histogram[J].International Journal of Image Processing,2010,3(6):282-293.
[22]Bhogal A K,Singla N,Kaur M.Color image segmentation based on color and texture properties[J].International Journal of Advanced Engineering Sciences and Technologies,2011,8(2):152-159.
[23]李凯,张建华,冯全,等.复杂背景与天气条件下的棉花叶片图像分割方法[J].中国农业大学学报,2018,23(2):88-98.Li Kai,Zhang Jianhua,Feng Quan,et al.Image segmentation method complex background and for cotton leaf under weather conditions[J].Journal of China Agricultural University,2018,23(2):88-98.(in Chinese with English abstract)
[24]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.Recognition of honey pomelo leaf diseases based on optimal binary tree support vector machine[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2014,30(19):222-231.(in Chinese with English abstract)
[25]江海洋,张建,袁媛,等.基于MDMP-LSM算法的黄瓜叶片病斑分割方法[J].农业工程学报,2012,28(21):142-148.Jiang Haiyang,Zhang Jian,Yuan Yuan,et al.Segmentation of cucumber disease leaf image based on MDMP-LSM[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2012,28(21):142-148.(in Chinese with English abstract)
[26]Ren Y G,Jian Z,Miao L,et al.Segmentation method for crop disease leaf images based on watershed algorithm[J].Journal of Computer Applications,2012,32(3):752-755.
[27]Liu T,Chen W,Wu W,et al.Detection of aphids in wheat fields using a computer vision technique[J].Biosystems Engineering,2016,141(11):82-93.
[28]王娅.血液红细胞图像自适应标记分水岭分割算法[J].中国图象图形学报,2017,22(12):1779-1787.Wang Ya.Adaptive marked watershed segmentation algorithm for red blood cell images[J].Journal of Image&Graphics,2017,22(12):1779-1787.(in Chinese with English abstract)
[29]方红萍,方康玲,刘新海,等.自适应H-minima的改进分水岭堆叠细胞分割方法[J].计算机应用研究,2016,33(5):1587-1590.Fang Hongping,Fang Kangling,Liu Xinhai,et al.Clustered cells segmentation using modified watershed method based on adaptive H-minima[J].Application Research of Computers,2016,33(5):1587-1590.(in Chinese with English abstract)
[30]石际亮,董黎君,梁国星.基于MATLAB图像处理的圆度误差数据采集方法[J].中国农机化学报,2015,36(2):257-261.Shi Jiliang,Dong Lijun,Liang Guoxing.Acquisition method of roundness error data based on MATLAB image processing[J].Journal of Chinese Agricultural Mechanization,2015,36(2):257-261.(in Chinese with English abstract)
[31]岳奎.最小二乘圆法评定圆度误差的程序设计[J].工具技术,2006,40(4):79-81.Yue Kui.Programming for evaluation of roundness error by least squares mean circle method[J].Tool Engineering,2006,40(4):79-81.(in Chinese with English abstract)
[32]Wang J,He J,Han Y,et al.An adaptive thresholding algorithm of field leaf image[J].Computers and Electronics in Agriculture,2013,96(6):23-39.