Computer Vision System for Railway Track Crack Detection using Deep Learning Neural Network
R. Thendral, A. Ranjeeth
Abstract
For better inspections and security, we need an efficient railway track crack detection system. In this research, we present a computer vision-based technique to detect the railway track cracks automatically. This system uses images captured by a rolling camera attached just below a self-moving vehicle in the railway department. The source images considered are the cracked and crack-free images. The first step is pre-processing scheme and then apply Gabor transform. In this paper, first order statistical features are extracted from the Gabor magnitude image. These extracted features are given as input to the deep learning neural network for differentiate the cracked track image from the non-cracked track image. Accuracy of the proposed algorithm on the procured images is 94.9 % and an overall error rate of 1.5%.