Research On Automatic Detection Of Steady Arm In Contactless Overhead Contact System

With the continuous development of the railway system, and rapid increase of the train speed, more funds are required to maintain the railway system. Meanwhile, maintaining the ancillary facilities of the electric locomotives has become an important issue.In this dissertation, we have focused on the research of detecting the Steady Arm in contactless overhead contact, and propose a solution based on machine vision which detects Steady Arm automatically. This method reduces the maintaining cost by improving the maintaining efficiency, and most importantly, minimizes manual mistakes.This system consists of a rapid image acquisition module, an image pre-processing module and a detection module. The image acquisition module contains an industrial-grade camera, which can accurately acquire images with high resolution. The image pre-processing module can obtain the contours of image effectively using Sobel edge operator. Then, the detection module calculates the rough angle of the Steady Arm by using the Chain Code detection algorithm followed by accurate localization using Radon Transform. This system can detect the angle of Steady Arm rapidly and accurately because of combining the efficiency of the Chain Code algorithm and the accuracy of Radon Transform.