Railway Track Fault Detection

Abstract

The digital nature of the data collection involved with a computer vision based method, archiving inspection results and trending of the data becomes feasible, leading to more advanced failure prediction models for maintenance scheduling and a more thorough understanding of railway track structure. Here a computer vision based method is presented. For better inspections and security, we need an efficient railway track crack detection system. In this project, 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 image processing. First order statistical features are extracted from the image processing and 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.