融合模糊增强与改进Canny的植物叶脉边缘提取
|
摘要
针对植物叶脉边缘提取存在的问题,提出一种图像模糊增强与改进Canny算子相结合的叶脉边缘提取方法。该方法首先构造分段模糊隶属度函数对高低灰度进行不同处理,然后通过模糊增强函数来增强叶脉两侧邻接区域灰度对比度,最后对增强图像采用改进Canny算子提取叶脉边缘。改进Canny算子根据像素点与局部像素的均值差来自适应的应用高斯滤波,以减小滤波对边缘的平滑影响;采用Prewitt模板计算梯度幅值和方向,以减小噪声对梯度计算的影响;采用线性插值技术实现全向非极大值抑制,以更准确的实现边缘细化。实验结果表明,本文方法在低照度叶片和光照不均叶片下提取叶脉边缘的PSNR分别为51.136 2dB和50.217 4dB,与传统边缘提取方法及相关文献相比,本文方法对边界模糊性不敏感,可以更加有效地提取完整精细的叶脉边缘。
Aiming at the problems existing in the extraction of plant leaf vein edges,a leaf vein edge extraction method combining image fuzzy enhancement and improved Canny operator is proposed.Firstly,apiecewise fuzzy membership function is constructed to perform different processing on the high and low gray value regions.Then,the gray contrast of the adjacent areas of the leaf veins is enhanced by fuzzy enhancement function.Finally,the leaf vein edges of the enhanced image are extracted by the improved Canny operator.The improved Canny operator adaptively applies Gaussian filtering according to the mean difference between pixels and local pixels to reduce the smoothing effect of filtering on the edges.The Prewitt template is used to calculate the gradient magnitude and direction to reduce the effect of noise on the gradient calculation.The linear interpolation technique is used to realize omnidirectional non-maximal suppression,and the edge refinement is achieved more accurately.The experimental results show that PSNR of the method for extracting leaf vein edges under low illumination and uneven illumination are51.1362 dB and 50.2174 dB,respectively.Compared with the traditional edge extraction method and related literatures,the method is not sensitive to the boundary ambiguity,and it can more effectively extract complete leaf vein edges.
Aiming at the problems existing in the extraction of plant leaf vein edges,a leaf vein edge extraction method combining image fuzzy enhancement and improved Canny operator is proposed.Firstly,apiecewise fuzzy membership function is constructed to perform different processing on the high and low gray value regions.Then,the gray contrast of the adjacent areas of the leaf veins is enhanced by fuzzy enhancement function.Finally,the leaf vein edges of the enhanced image are extracted by the improved Canny operator.The improved Canny operator adaptively applies Gaussian filtering according to the mean difference between pixels and local pixels to reduce the smoothing effect of filtering on the edges.The Prewitt template is used to calculate the gradient magnitude and direction to reduce the effect of noise on the gradient calculation.The linear interpolation technique is used to realize omnidirectional non-maximal suppression,and the edge refinement is achieved more accurately.The experimental results show that PSNR of the method for extracting leaf vein edges under low illumination and uneven illumination are51.1362 dB and 50.2174 dB,respectively.Compared with the traditional edge extraction method and related literatures,the method is not sensitive to the boundary ambiguity,and it can more effectively extract complete leaf vein edges.
引文
[1] Larese M G,BayáA E,Craviotto R M,et al.Multiscale recognition of legume varieties based on leaf venation images[J].Expert Systems with Applications,2014,41(10):4638-4647.
[2] WANG Xian-feng,ZHANG Shan-wen,WANG Zhen,et al.Recognition of cucumber diseases based on leaf image and environmental information[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2014,30(14):148-153.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.
[3] Singh V,Misra A K.Detection of plant leaf diseases using image segmentation and soft computing techniques[J].Information Processing in Agriculture,2017,4(1):41-49.
[4] Kiran R.Gavhale,Ujwalla G.An overview of the research on plant leaves disease detection using image processing techniques[J].IOSR Journal of Computer Engineering(IOSR-JCE),2014,16(1):10-16.
[5] HE Jin,WU Cheng-mao.Kernel space adaptive non-local means robust segmentation algorithm[J].Journal of Optoelectronics·Laser,2017,28(8):910-917.何晶,吴成茂.核空间自适应非局部均值鲁棒分割算法[J].光电子·激光,2017,28(8):910-917.
[6] JIAO Ren-zhi,CHEN Fen,WANG Hui,et al.A depth video rectification algorithm based on image segmentation[J].Journal of Optoelectronics·Laser,2016,27(1):97-105.焦任直,陈芬,汪辉,等.基于图像分割的深度视频校正算法[J].光电子·激光,2016,27(1):97-105.
[7] LEI Kai-yun,QIN Xun-peng,LIU Hua-ming,et al.Weld pool edge extraction in wild-band laser cladding based on local region active contour model[J].Journal of Optoelectronics·Laser,2018,29(5):516-522.雷凯云,秦训鹏,刘华明,等.基于局部主动轮廓模型的宽带激光熔覆熔池边缘提取[J].光电子·激光,2018,29(5):516-522.
[8] LI Jun-feng,ZHANG Zhi-xiang,SHEN Jun-min.Study on surface defect extraction of magnetic ring based on masking technology[J].Journal of Optoelectronics·Laser,2017,28(7):732-741.李俊峰,张之祥,沈军民.基于掩模图像的磁环表面缺陷提取研究[J].光电子·激光,2017,28(7):732-741.
[9] FU Hong,CHI Zhe-ru,CHANG Jie,et al.Extraction of leaf vein based on artificial neural network:studies on the living plant identificationⅠ[J].Chinese Bulletin of Botany,2004,21(4):429-436.傅弘,池哲儒,常杰,等.基于人工神经网络的叶脉信息提取-植物活体机器识别研究Ⅰ[J].植物学通报,2004,21(4):429-436.
[10] Wang Z,Sun X,Zhu Y,et al.Leaf recognition based on PCNN[J].Neural Computing&Applications,2016,27(4):899-908.
[11]Xiaodong Z,Xiaojie W.Leaf vein extraction based on gray-scale morphology[J].International Journal of Image Graphics&Signal Processing,2010,2(2):25-31.
[12]Rahmadhani M,Yeni Herdiyeni.Shape and vein extraction on plant leaf images using Fourier and B-spline modeling[C].AFITA 2010 International Conference,The Quality Information for Competitive Agricultural Based Production System and Commerce,2010,475-490.
[13]Katyal V,Aviral.Leaf vein segmentation using odd Gabor filters and morphological operations[J].International Journal of Advanced Research in Computer Science,2012,3(3):13-18.
[14] Xiaodong Z,Xiaojie W.Fast leaf vein extraction using hue and intensity information[C].International Conference on Information Engineering and Computer Science,IEEE,2009,1-4.
[15]LI Can-can,SUN Chang-hui,WANG Jing,et al.Extraction of leaf vein based on improved Sobel algorithm and hue Information[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2011,27(7):196-199.李灿灿,孙长辉,王静,等.基于改进的Sobel算子和色调信息的叶脉提取方法[J].农业工程学报,2011,27(7):196-199.
[16]Pal S K,King R A.On edge detection of X-ray images using fuzzy sets[J].Pattern Analysis&Machine Intelligence IEEE Transactions on,1983,5(1):69-77.
[17]Liu X.An improved image enhancement algorithm based on fuzzy set[J].Physics Procedia,2012,33:790-797.
[18]XU Yan-lei,JIA Hong-lei,BAO Jia-lin.Plant leaf vein edge detection based on fuzzy order morphology[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(13):193-198.徐艳蕾,贾洪雷,包佳林.基于模糊顺序形态学的植物叶片脉络边缘提取[J].农业工程学报,2015,31(13):193-198.
[19]LIN Kai-yan,SI Hui-ping,ZHOU Qiang,et al.Plant leaf edge detection based on fuzzy logic[J].Transactions of the Chinese Society for Agricultural Machinery,2013,44(6):227-231.林开颜,司慧萍,周强,等.基于模糊逻辑的植物叶片边缘检测方法[J].农业机械学报,2013,44(6):227-231.
[20] LI Can-can,WANG Bao,WANG Jing,et al.Extracting vein of leaf image based on K-means clustering[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2012,28(17):157-162.李灿灿,王宝,王静,等.基于K-means聚类的植物叶片图像叶脉提取[J].农业工程学报,2012,28(17):157-162.
[21] LIU Tao-sheng,LI Pei-hong,LI Chen-feng.Technology research on RS image self-adaption linear enhancement combined with pixel gray frequency distribution[J].Journal of Jiangxi University of Science and Technology,2014,35(5):40-44.刘陶胜,李沛鸿,李辰风.结合像素频率分布特征的遥感图像自适应线性增强[J].江西理工大学学报,2014,35(5):40-44.
[22] YIN Shi-bai,KONG Chui-han,WANG Yi-bin.Unsupervised hierarchical human target infrared image segmentation through fuzzy correlation[J].Optics and Precision Engineering,2018,26(6):1542-1550.尹诗白,孔垂涵,王一斌.非监督层次化模糊相关的人体红外图像分割[J].光学精密工程,2018,26(6):1542-1550.
[2] WANG Xian-feng,ZHANG Shan-wen,WANG Zhen,et al.Recognition of cucumber diseases based on leaf image and environmental information[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2014,30(14):148-153.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.
[3] Singh V,Misra A K.Detection of plant leaf diseases using image segmentation and soft computing techniques[J].Information Processing in Agriculture,2017,4(1):41-49.
[4] Kiran R.Gavhale,Ujwalla G.An overview of the research on plant leaves disease detection using image processing techniques[J].IOSR Journal of Computer Engineering(IOSR-JCE),2014,16(1):10-16.
[5] HE Jin,WU Cheng-mao.Kernel space adaptive non-local means robust segmentation algorithm[J].Journal of Optoelectronics·Laser,2017,28(8):910-917.何晶,吴成茂.核空间自适应非局部均值鲁棒分割算法[J].光电子·激光,2017,28(8):910-917.
[6] JIAO Ren-zhi,CHEN Fen,WANG Hui,et al.A depth video rectification algorithm based on image segmentation[J].Journal of Optoelectronics·Laser,2016,27(1):97-105.焦任直,陈芬,汪辉,等.基于图像分割的深度视频校正算法[J].光电子·激光,2016,27(1):97-105.
[7] LEI Kai-yun,QIN Xun-peng,LIU Hua-ming,et al.Weld pool edge extraction in wild-band laser cladding based on local region active contour model[J].Journal of Optoelectronics·Laser,2018,29(5):516-522.雷凯云,秦训鹏,刘华明,等.基于局部主动轮廓模型的宽带激光熔覆熔池边缘提取[J].光电子·激光,2018,29(5):516-522.
[8] LI Jun-feng,ZHANG Zhi-xiang,SHEN Jun-min.Study on surface defect extraction of magnetic ring based on masking technology[J].Journal of Optoelectronics·Laser,2017,28(7):732-741.李俊峰,张之祥,沈军民.基于掩模图像的磁环表面缺陷提取研究[J].光电子·激光,2017,28(7):732-741.
[9] FU Hong,CHI Zhe-ru,CHANG Jie,et al.Extraction of leaf vein based on artificial neural network:studies on the living plant identificationⅠ[J].Chinese Bulletin of Botany,2004,21(4):429-436.傅弘,池哲儒,常杰,等.基于人工神经网络的叶脉信息提取-植物活体机器识别研究Ⅰ[J].植物学通报,2004,21(4):429-436.
[10] Wang Z,Sun X,Zhu Y,et al.Leaf recognition based on PCNN[J].Neural Computing&Applications,2016,27(4):899-908.
[11]Xiaodong Z,Xiaojie W.Leaf vein extraction based on gray-scale morphology[J].International Journal of Image Graphics&Signal Processing,2010,2(2):25-31.
[12]Rahmadhani M,Yeni Herdiyeni.Shape and vein extraction on plant leaf images using Fourier and B-spline modeling[C].AFITA 2010 International Conference,The Quality Information for Competitive Agricultural Based Production System and Commerce,2010,475-490.
[13]Katyal V,Aviral.Leaf vein segmentation using odd Gabor filters and morphological operations[J].International Journal of Advanced Research in Computer Science,2012,3(3):13-18.
[14] Xiaodong Z,Xiaojie W.Fast leaf vein extraction using hue and intensity information[C].International Conference on Information Engineering and Computer Science,IEEE,2009,1-4.
[15]LI Can-can,SUN Chang-hui,WANG Jing,et al.Extraction of leaf vein based on improved Sobel algorithm and hue Information[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2011,27(7):196-199.李灿灿,孙长辉,王静,等.基于改进的Sobel算子和色调信息的叶脉提取方法[J].农业工程学报,2011,27(7):196-199.
[16]Pal S K,King R A.On edge detection of X-ray images using fuzzy sets[J].Pattern Analysis&Machine Intelligence IEEE Transactions on,1983,5(1):69-77.
[17]Liu X.An improved image enhancement algorithm based on fuzzy set[J].Physics Procedia,2012,33:790-797.
[18]XU Yan-lei,JIA Hong-lei,BAO Jia-lin.Plant leaf vein edge detection based on fuzzy order morphology[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2015,31(13):193-198.徐艳蕾,贾洪雷,包佳林.基于模糊顺序形态学的植物叶片脉络边缘提取[J].农业工程学报,2015,31(13):193-198.
[19]LIN Kai-yan,SI Hui-ping,ZHOU Qiang,et al.Plant leaf edge detection based on fuzzy logic[J].Transactions of the Chinese Society for Agricultural Machinery,2013,44(6):227-231.林开颜,司慧萍,周强,等.基于模糊逻辑的植物叶片边缘检测方法[J].农业机械学报,2013,44(6):227-231.
[20] LI Can-can,WANG Bao,WANG Jing,et al.Extracting vein of leaf image based on K-means clustering[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2012,28(17):157-162.李灿灿,王宝,王静,等.基于K-means聚类的植物叶片图像叶脉提取[J].农业工程学报,2012,28(17):157-162.
[21] LIU Tao-sheng,LI Pei-hong,LI Chen-feng.Technology research on RS image self-adaption linear enhancement combined with pixel gray frequency distribution[J].Journal of Jiangxi University of Science and Technology,2014,35(5):40-44.刘陶胜,李沛鸿,李辰风.结合像素频率分布特征的遥感图像自适应线性增强[J].江西理工大学学报,2014,35(5):40-44.
[22] YIN Shi-bai,KONG Chui-han,WANG Yi-bin.Unsupervised hierarchical human target infrared image segmentation through fuzzy correlation[J].Optics and Precision Engineering,2018,26(6):1542-1550.尹诗白,孔垂涵,王一斌.非监督层次化模糊相关的人体红外图像分割[J].光学精密工程,2018,26(6):1542-1550.