基于高斯尺度空间和SVM的桥梁裂缝检测研究
|
摘要
针对现有的桥梁裂缝检测及分类算法在光照不均匀条件下,存在检测精度不高、分类效果不理想的问题,提出了一种基于高斯尺度空间与支持向量机(sopport vector machine,SVM)多分类器相结合的桥梁裂缝检测及分类算法。该文对待处理裂缝图像进行预处理,消除噪声干扰;通过裂缝图像与二维高斯函数进行卷积运算来创建高斯尺度空间,在高斯尺度空间下估计背景,利用背景差法消除光照干扰,进而提取出具有区分度的裂缝图像特征向量;利用SVM多分类器进行桥梁裂缝分类。实验结果表明所提出的算法具有较高的分类精度。
Aiming at the problem that the existing bridge crack detection and classification algorithm has low detection accuracy and unsatisfactory classification under the condition of uneven illumination,a Gaussian scale space and sopport vector machine( SVM) is proposed. Bridge crack detection and classification algorithm combined with classifier. In this paper,the processed crack image is preprocessed to eliminate noise interference. The Gaussian scale space is created by the convolution operation of the crack image and the two-dimensional Gaussian function. The background is estimated in the Gaussian scale space,and the illumination interference is eliminated by the background difference method. The crack image feature vector with discriminant degree is obtained. The SVM multi-classifier is used to classify the crack of the bridge. Experimental results show that the proposed algorithm has higher classification accuracy.
Aiming at the problem that the existing bridge crack detection and classification algorithm has low detection accuracy and unsatisfactory classification under the condition of uneven illumination,a Gaussian scale space and sopport vector machine( SVM) is proposed. Bridge crack detection and classification algorithm combined with classifier. In this paper,the processed crack image is preprocessed to eliminate noise interference. The Gaussian scale space is created by the convolution operation of the crack image and the two-dimensional Gaussian function. The background is estimated in the Gaussian scale space,and the illumination interference is eliminated by the background difference method. The crack image feature vector with discriminant degree is obtained. The SVM multi-classifier is used to classify the crack of the bridge. Experimental results show that the proposed algorithm has higher classification accuracy.
引文
[1]董安国,宋君,张仙艳,等.基于图像的桥梁裂缝检测算法[J].自动化仪表,2013(8):1-5.
[2]姚渊.便携式桥梁检测系统的设计与实现[D].西安:长安大学,2016.
[3] Postema F,Van B A. NDT used in the Netherlands from a principal point of view[C]//Berlin:Proceeding of the International Symposium of Non-destructive Testing in Civil Engineering,2003.
[4]宋君.基于数字图像的混凝土裂缝检测算法研究[D].西安:长安大学:2013:1-34.
[5] Zhang Jingjing,Nie Hongyu,Yu Qiang. Bridge Crack Detection Based on Percolation Model with Multi Scale Input Image[J]. Computer Engineering,2017,43(2):273-279.
[6]陈瑶,梅涛,王晓杰,等.基于爬壁机器人的桥梁裂缝图像检测与分类方法[J].中国科学技术大学学报,2016,46(9):788-796.
[7]魏武,王俊杰,蔡钊雄.基于小波和Radon变换的桥梁裂缝检测[J].计算机工程与设计,2013,34(9):3151-3156.
[8]龙建武,申铉京,臧慧,等.高斯尺度空间下估计背景的自适应阈值分割算法[J].自动化学报,2014,40(8):1773-1782.
[9] Shen Xuan-Jing,Long Jian-Wu,Chen Hai-Peng,et al. Otsu thresholding algorithm based on rebuilding and dimension reduction of the 3-dimensional histogram[J].Acta Electronica Sinica,2011,38(5):1108-1114.
[10] Krinidis S,Chatzis V. A robust fuzzy local information C means clustering algorithm[J]. IEEE Transactions on Image Processing,2010,19(5):1328-1337.
[11]王陈飞,肖诗斌.基于SVM的图像分类研究[J].计算机与数字工程,2006,34(8):74-75.
[2]姚渊.便携式桥梁检测系统的设计与实现[D].西安:长安大学,2016.
[3] Postema F,Van B A. NDT used in the Netherlands from a principal point of view[C]//Berlin:Proceeding of the International Symposium of Non-destructive Testing in Civil Engineering,2003.
[4]宋君.基于数字图像的混凝土裂缝检测算法研究[D].西安:长安大学:2013:1-34.
[5] Zhang Jingjing,Nie Hongyu,Yu Qiang. Bridge Crack Detection Based on Percolation Model with Multi Scale Input Image[J]. Computer Engineering,2017,43(2):273-279.
[6]陈瑶,梅涛,王晓杰,等.基于爬壁机器人的桥梁裂缝图像检测与分类方法[J].中国科学技术大学学报,2016,46(9):788-796.
[7]魏武,王俊杰,蔡钊雄.基于小波和Radon变换的桥梁裂缝检测[J].计算机工程与设计,2013,34(9):3151-3156.
[8]龙建武,申铉京,臧慧,等.高斯尺度空间下估计背景的自适应阈值分割算法[J].自动化学报,2014,40(8):1773-1782.
[9] Shen Xuan-Jing,Long Jian-Wu,Chen Hai-Peng,et al. Otsu thresholding algorithm based on rebuilding and dimension reduction of the 3-dimensional histogram[J].Acta Electronica Sinica,2011,38(5):1108-1114.
[10] Krinidis S,Chatzis V. A robust fuzzy local information C means clustering algorithm[J]. IEEE Transactions on Image Processing,2010,19(5):1328-1337.
[11]王陈飞,肖诗斌.基于SVM的图像分类研究[J].计算机与数字工程,2006,34(8):74-75.