Vibration compensation for railway track displacement monitoring system using biomedical image processing concept

Abstract

Vibration compensation in track displacement monitoring is explored by using rail image feature points. The key point is to compare the image of the rail to be tested with the image of the standard rail. A downstream section of Metro Line A is selected for real time analysis of image. The sampling frequency of the detection system is 0.5m/point, and the driving speed of the detection vehicle is about 40km/h, take the distance of 100m between K23+000-K22+900 of kilometer mark, total 200 points, and analyze the image data by grey scale conversion concept. In order to verify the accuracy of the vibration compensation data, one point is taken at an interval of 5m, and a total of 20 points are measured, and compare and analyze with dynamic real-time detection data. After vibration compensation, the geometric parameters of the contact rail are closer to the expected value, and the maximum error can be controlled within ±3mm. Using biomedical image-based processing concept to realize the track displacement monitoring method, which meets the requirements of non-contact measurement, and the degree of automation is high, it can be monitored and alarmed in real time, and it has a good application prospect.