[1] [1] Magel E, Sroba P, Sawley K, et al. Control of rolling contactfatigue of rails[J]. Center for Surface Transportation Technol.ogy, National Research Council Canada, 2005.

[6] [6] Gonzalez, R C, Woods R E. Digital Image Processing, Read.ing MA, USA: Addison Wesley, 1993.

[7] [7] Deutschl E, Gasser C, Niel A, et al. Defect detection on railsurfaces by a vision based system[C]//Intelligent VehiclesSymposium, 2004 IEEE. IEEE, 2004: 507-511.

[8] [8] Mandriota C, Nitti M, Ancona N, et al. Filter-based featureselection for rail defect detection [J]. Machine Vision andApplications, 2004, 15(4): 179-185.

[9] [9] Dare P, Hanley H, Fraser C, et al. An operational applica.tion of automatic feature extraction: the measurement of cracks in concrete structures [J]. The Photogrammetric Re.cord, 2002, 17(99): 453-464.

[10] [10] Yamaguchi T, Hashimoto S. Fast crack detection method forlarge-size concrete surface images using percolation-based image processing [J]. Machine Vision and Applications,2010, 21(5): 797-809.

[11] [11] Yamaguchi T, Hashimoto S. Improved percolation-based method for crack detection in concrete surface images[C]//Pattern Recognition, 2008. ICPR 2008. 19th InternationalConference on IEEE, 2008: 1-4.