Aiming at the high speed sliding electrical contact rail, the surface micro damage detection and identification method are studied. Based on the principle of laser scanning, a three-dimensional measurement system is constructed for acquisition of point cloud information for rail surface, meanwhile a method of depth mapping color based on point cloud is proposed for the detection of rail surface micro damage. After the threedimensional point cloud data being denoised, smoothed and reduced, according to the datum plane set, point cloud depth mapping color model is constructed, the point cloud depth information can be mapped into red, green, blue (RGB) information, and the optimum color threshold is set by using the one-dimensional maximum entropy method to realize the accurate extraction of the damage area. Binary tree pattern recognition method is used to establish damage classification model and realize the identification and classification of rail surface micro damage. The results show that detection rate of the micro damage of which mass loss is less than 1 gram is more than ninety-eight percent, detection accuracy of minor mass loss is milligram and the damage identification rate of pits and scratches can reach above eighty percent.