绝缘子陶瓷裂缝识别的Sobel改进算法
|
摘要
为准确快速识别绝缘子裂缝,在图像处理的基础上,提出了一种基于改进Sobel算法的绝缘子陶瓷裂缝识别方法。先对采集到的图像进行预处理,减弱目标裂缝的噪声干扰,利用改进的Sobel算法对裂缝进行边缘提取,并采用最大类间方差法将含有裂缝边缘信息的图像数据的前景与背景分开,从而提取出完整的绝缘子裂缝图像。结果表明:改进后的Sobel算法检测的裂缝边缘更加完整和连续,裂缝图像线条更流畅清晰,产生的毛刺较少,抗噪声干扰能力得到增强。该研究为后期更为精准地提取绝缘子裂缝的面积、长度等具体特征值提供了有效保障。
This paper proposes an insulator ceramic crack recognition method based on improved Sobel algorithm on the basis of image processing,an approach capable of accurately and quickly identifying insulator cracks. The targeted study involves reducing the noise interference of the target crack by preprocessing the collected images; extracting the edge of the crack using the improved Sobel algorithm; separating the foreground and background of the image data containing the edge information of the crack using the method of maximum inter-class variance; and thereby extracting the complete insulator crack image.The results show that the improved Sobel algorithm enables detection of more complete and continuous fracture edges,smoother and clearer fracture image lines,less burrs and enhanced anti-noise interference. This study could provide an effective guarantee for more accurate extraction of specific characteristic values such as the area and length of insulator cracks in the later stage.
This paper proposes an insulator ceramic crack recognition method based on improved Sobel algorithm on the basis of image processing,an approach capable of accurately and quickly identifying insulator cracks. The targeted study involves reducing the noise interference of the target crack by preprocessing the collected images; extracting the edge of the crack using the improved Sobel algorithm; separating the foreground and background of the image data containing the edge information of the crack using the method of maximum inter-class variance; and thereby extracting the complete insulator crack image.The results show that the improved Sobel algorithm enables detection of more complete and continuous fracture edges,smoother and clearer fracture image lines,less burrs and enhanced anti-noise interference. This study could provide an effective guarantee for more accurate extraction of specific characteristic values such as the area and length of insulator cracks in the later stage.
引文
[1]杨璐雅,黄新波,张烨,等.基于边缘检测的瓷质绝缘子裂缝检测方法[J].广东电力,2018,7(31):106-111.
[2]蒋志佳.图像处理技术在架空线路绝缘子裂缝识别中的应用作用研究[D].江苏:江苏大学,2009:1-15.
[3]王章锋.边缘检测方法研究及应用[D].天津:天津工业大学,2017:15-28.
[4]唐阳山,徐忠帅,黄贤丞,等.基于Roberts算子的车道线图像的边缘检测研究[J].辽宁工业大学学报(自然科学版),2017,37(6):383-390.
[5]虎玲,常霞,纪峰.图像边缘检测方法研究新进展[J].现代电子技术,2018,41(23):32-37.
[6]李洪安,张飞,杜卓明.针对合成孔径雷达图像的新型LOG边缘检测算法[D].西安:西安科技大学,2014:1-15.
[7]张春燕.基于边缘检测经典算法的改进研究与实现[D].合肥:安徽大学,2012:4-17.
[8]Alam M A,Ali M N,Syed A A,et al.An algorithm to detect and identify defects of industrial pipes using image processing[C]//Proceedings of the 8th International Conference on Software,Knowledge Inf-omation Management and Applications.Dhaka,Bangladesh,2015:1-6.
[9]胡皙.基于图像处理的地铁隧道裂缝检测技术研究[D].北京:北京交通大学,2014:1-10.
[10]Khalifa I,Aboutabl A E,Barakat G S A.A new image-based model for predicting cracks in sewer pipes[J].International journal of advanced computer science and applications,2013,4(12):65-71.
[11]王聪雅.基于图像识别处理的桥梁底面裂缝检测方法的研究[D].北京:北京交通大学,2016:15-33.
[12]Sanjay S,Anil K S,Ravi S,et al.A novel real-ti meresource efficient implementation of Sobel operator-based edge detection on FPGA[J].International Journalof Electronics,2014,101(12):1705-1715.
[13]沈德海,侯建,鄂旭.基于改进的Sobel算子边缘检测算法[J].计算机技术与发展,2013,23(11):22-25.
[14]Gongzalez C I,Mflin P,Castro J R,et al.An improved sobcl edge detection method based on generalized type-2 fuzzy logic[J].Soft Comput,2016,20(2):773-784.
[15]Bo P.Bohai front detection based on multi-scale Sobel algorithm[J].IEEE Geoscience and Remote Sensing Symposium,2014,17(1):4423-4426.
[2]蒋志佳.图像处理技术在架空线路绝缘子裂缝识别中的应用作用研究[D].江苏:江苏大学,2009:1-15.
[3]王章锋.边缘检测方法研究及应用[D].天津:天津工业大学,2017:15-28.
[4]唐阳山,徐忠帅,黄贤丞,等.基于Roberts算子的车道线图像的边缘检测研究[J].辽宁工业大学学报(自然科学版),2017,37(6):383-390.
[5]虎玲,常霞,纪峰.图像边缘检测方法研究新进展[J].现代电子技术,2018,41(23):32-37.
[6]李洪安,张飞,杜卓明.针对合成孔径雷达图像的新型LOG边缘检测算法[D].西安:西安科技大学,2014:1-15.
[7]张春燕.基于边缘检测经典算法的改进研究与实现[D].合肥:安徽大学,2012:4-17.
[8]Alam M A,Ali M N,Syed A A,et al.An algorithm to detect and identify defects of industrial pipes using image processing[C]//Proceedings of the 8th International Conference on Software,Knowledge Inf-omation Management and Applications.Dhaka,Bangladesh,2015:1-6.
[9]胡皙.基于图像处理的地铁隧道裂缝检测技术研究[D].北京:北京交通大学,2014:1-10.
[10]Khalifa I,Aboutabl A E,Barakat G S A.A new image-based model for predicting cracks in sewer pipes[J].International journal of advanced computer science and applications,2013,4(12):65-71.
[11]王聪雅.基于图像识别处理的桥梁底面裂缝检测方法的研究[D].北京:北京交通大学,2016:15-33.
[12]Sanjay S,Anil K S,Ravi S,et al.A novel real-ti meresource efficient implementation of Sobel operator-based edge detection on FPGA[J].International Journalof Electronics,2014,101(12):1705-1715.
[13]沈德海,侯建,鄂旭.基于改进的Sobel算子边缘检测算法[J].计算机技术与发展,2013,23(11):22-25.
[14]Gongzalez C I,Mflin P,Castro J R,et al.An improved sobcl edge detection method based on generalized type-2 fuzzy logic[J].Soft Comput,2016,20(2):773-784.
[15]Bo P.Bohai front detection based on multi-scale Sobel algorithm[J].IEEE Geoscience and Remote Sensing Symposium,2014,17(1):4423-4426.