Research on Inverse P-M Diffusion-based Rail Surface Defect Detection
-
摘要: 研制了一种基于反向P-M(Perona-Malik)扩散的钢轨表面缺陷视觉检测装置,该装置可 自动获取钢轨表面图像,并实现实时检测与定位钢轨表面缺陷. 钢轨图像具有光 照变化、反射不均、特征少等特点,为了在运动过程中 从复杂的钢轨表面图像提取缺陷,首先将图像进行反向P-M扩散,然后将扩散后的图像与原图像进 行差分,从而减小了上述因素的影响,最后将差分图像进行二值化操作,根据 缺陷边缘特性和面积进行滤波,分割出缺陷图像. 实验仿真和现场测试结果表明,该方法能很好地识别块状缺陷和线状缺陷,并且检测速度、精度、识别 率和误检率都能很好地满足要求.Abstract: A vision machine is developed for rail surface defects detection based on the inverse P-M (Perona-Malik) diffusion. The rail surface defects images can be obtained through an image acquisition system. The rail surface images show illumination variation, reflection inequality, and heterogeneous texture, they make the automated visual inspection task extremely difficult. The faultless region of the rail surface image is preserved by an inverse P-M model, but the fault region is smoothed after diffusing by an inverse P-M model. Therefore, by subtracting the inverse diffused image from the original image, the defects can be distinctly enhanced in the difference image. The influence of illumination variation, reflection inequality, and heterogeneous texture can also be decreased. A simple binary thresholding, followed by filter operations based on the edge performance and the size of defects, can then easily segment the defect. The simulation and field experiments indicate that the inspection machine can detect the rail surface defects effectively and the detection speed, accuracy, detection ratio and the fault ratio also satisfy the needs of automated rail track.
-
[1] Jasiūnien E, žukauskas E. The ultrasonic wave interaction with porosity defects in welded rail head. ULTRAGARSAS (ULTRASOUND), 2010, 65(1): 12-18 [2] [2] Vidaud M, Zwanenburg W J. Current situation on rolling contact fatiguea rail wear phenomenon. In: Proceedings of the 9th Swiss Transport Research Conference. Monte Verit, Swiss, 2009. 1-27 [3] [3] Marino F, Distante A, Mazzeo P L, Stella E. A real-time visual inspection system for railway maintenance: automatic hexagonal-headed bolts. IEEE Transactions on Systems Man, and Cybernetics, Part C: Applications and Reviews, 2007, 37(3): 418-428 [4] [4] Mandriota C, Stella E, Nitti M, Ancona N, Distante A. Rail corrugation detection by Gabor filtering. In: Proceedings of IEEE International Conference on Image Processing. Thessaloniki: IEEE, 2001. 626-628 [5] [5] Mandriota C, Nitti M, Ancona N, Stella E, Distante A. Filter-based feature selection for rail defect detection. Machine Vision and Applications, 2004, 15(4): 179-185 [6] [6] Papaelias M P, Roberts C, Davis C L. A review on non-destructive evaluation of rails: state-of-the-art and future development. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and rapid transit, 2008, 222(4): 367-384 [7] [7] Deutschl E, Gasser C, Niel A, Werschonig J. Defect detection on rail surfaces by a vision based system. In: Proceedings of IEEE Intelligent Vehicles Symposium. Parma, Italy: IEEE, 2004. 507-511 [8] [8] Shah M. Automated Visual Inspection/Detection of Railroad Track, Technical Report, BD550-08, Computer Vision Lab, University of Central Florida, USA, 2010 [9] [9] Li Q Y, Ren S W. A real-time visual inspection system for discrete surface defects of rail heads. IEEE Transactions on Instrumentation and Measurement, 2012, 61(8): 2189-2199 [10] Xie X H. A review of recent advances in surface defect detection using texture analysis techniques. Electronic Letters on Computer Vision and Image Analysis, 2008, 7(3): 1-22 [11] Tsai D M, Chao S M. An anisotropic diffusion-based defect detection for sputtered surfaces with inhomogeneous textures. Image Vision Computing, 2005, 23(3): 325-338 [12] Tsai D M, Chang C C, Chao S M. Micro-crack inspection in heterogeneously textured solar wafers using anisotropic diffusion. Image Vision Computing, 2010, 28(3): 491-501 [13] Chao S M, Tsai D M. An anisotropic diffusion-based defect detection for low-contrast glass substrates. Image Vision Computing, 2008, 26(2): 187-200 [14] Chao S M, Tsai D M. Anisotropic diffusion with generalized diffusion coefficient function for defect detection in low-contrast surface images. Pattern Recognition, 2010, 43: 1917-1931 [15] Chao S M, Tsai D M. An improved anisotropic diffusion model for detail and edge-preserving smoothing. Pattern Recognition, 2010, 31(13): 2012-2023 [16] Chao S M, Tsai D M, Tseng Y H, Jhang Y R. Defect detection in low-contrast glass substrates using anisotropic diffusion. In: Proceedings of the 18th International Conference on Pattern Recognition. Hong Kong, China: IEEE, 2006. 654-657 [17] Chao S M, Tsai D M, Li W C, Chiu W Y. A generalized anisotropic diffusion for defect detection in low-contrast surfaces. In: Proceedings of the 20th International Conference on Pattern Recognition. Istanbul, Turkey: IEEE, 2010. 4408-4411 [18] Mr [19] zek P, Weickert J, Steidl G. Correspondences between wavelet shrinkage and nonlinear diffusion. Lecture Notes in Computer Science, 2003, 2695: 101-116 [20] Shih A C, Liao H M, Lu C S. A new iterated two-band diffusion equation-theory and its application. IEEE Transactions on Image Processing, 2003, 12(4): 466-476 [21] Yue Y, Croitoru M M, Bidani A, Zwischenberger J B, Clark J W Jr. Nonlinear multiscale wavelet diffusion for speckle suppression and edge enhancement in ultrasound images. IEEE Transactions on Medical Imaging, 2006, 25(3): 297-311 [22] Liu Wei-Wei, Yan Yun-Hui, Li Zhan-Yu, Li Jun. An image filtering algorithm for online detection system of steel strip surface defects. Journal of Northeastern University: Natural Science, 2009, 30(3): 430-433(刘伟嵬, 颜云辉, 李瞻宇, 李骏. 带钢表面缺陷在线检测系统的图像滤波算法. 东北大学学报: 自然科学版, 2009, 30(3): 430-433) [23] Magel E E. Rolling Contact Fatigue: a Comprehensive Review, Technical Report, DTFR53-05-H-00203, Centre for Surface Technology, Ottawa, Canada, 2011 [24] Weickert J. Anisotropic Diffusion in Image Processing. Stuttgart: Teubner-Verlag, 1998 [25] Li Xiao-Guang, Shen Lan-Sun, Lam Kin-Man, Wang Su-Yu. An image magnification method with GVF-based anisotropic diffusion model. Acta Electronica Sinica, 2008, 36(9): 1755-1758(李晓光, 沈兰荪, Lam Kin-Man, 王素玉. 一种基于GVF各向异性扩散模型的图像放大算法. 电子学报, 2008, 36(9): 1755-1758) [26] Meng Xiang-Lin, Wang Zheng-Zhi. Image diffusion based on visual masking effect. Acta Automatica Sinica, 2011, 37(1): 21-27(孟祥林, 王正志. 基于视觉掩蔽效应的图像扩散. 自动化学报, 2011, 37(1): 21-27) [27] Li Can-Fei, Wang Yao-Nan, Xiao Chang-Yan, Lu Xiao. A new speckle reducing anisotropic diffusion for ultrasonic speckle. Acta Automatica Sinica, 2012, 38(3): 412-419(李灿飞, 王耀南, 肖昌炎, 卢笑. 用于超声斑点噪声滤波的各向异性扩散新模型. 自动化学报, 2012, 38(3): 412-419) [28] Zheng Qiang, Dong En-Qing. Narrow band active contour model for local segmentation of medical and texture images. Acta Automatica Sinica, 2013, 39(1): 21-30(郑强, 董恩清. 窄带主动轮廓模型及在医学和纹理图像局部分割中的应用. 自动化学报, 2013, 39(1): 21-30) [29] Wang Zhi-Ming, Zhang Li. Local-structure-adapted image diffusion. Acta Automatica Sinica, 2009, 35(3): 244-250(王志明, 张丽. 局部结构自适应的图像扩散. 自动化学报, 2009, 35(3): 244-250) [30] Cheng Jian-Gang, Tian Jie, He Yu-Liang, Yang Xin. Fingerprint enhancement algorithm based on nonlinear diffusion filter. Acta Automatica Sinica, 2004, 30(6): 854-862(程建刚, 田捷, 何余良, 杨鑫. 基于非线性扩散滤波的指纹增强算法. 自动化学报, 2004, 30(6): 854-862) [31] Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(7): 629-639
点击查看大图
计量
- 文章访问数: 2292
- HTML全文浏览量: 113
- PDF下载量: 2257
- 被引次数: 0