Real-time rail head surface defect detection: A geometrical approach

Rail head surface defect detection is a major issue for rail maintenance, which is mainly used to avoid railway accidents due to rail track failures. The aim of this paper is to present a new vision based inspection technique for detecting special Rolling Contact Fatigue (RCF) defects that particularly occur on rail head surface, meanwhile, an automatic detecting system is implemented, which consists of pre-processing, defect locating, defect identifying and post-processing subsystems. To realize the defect locating sub-procedure, a simple and fast algorithm has been proposed, which adopts geometrical analysis directly on a gray-level histogram curve (the first-order statistical texture property) of the smoothed rail head surface image. Experimental results show that the proposed algorithm has a higher precision and is more suitable than the baseline method for real-time rail head surface defect detection application.