Rail profile matching method based on stepped calibration plate and improved SURF algorithm

Abstract

Abstract In order to avoid the calibration plane and laser plane do not coincide and the rail profile is difficult to match dynamically, a stepped calibration plate is made, and a rail profile matching method based on the improved SURF algorithm and random sample consensus algorithm (RANSAC) is proposed. Firstly, a stepped rail profile calibration plate is designed and fabricated, which can directly calibrate the laser plane, avoiding the problem that the calibration plane and the line laser plane do not coincide completely in practical application. In the calibration process, collect the optimal control points and calibrate the image by polynomial transformation and bilinear interpolation, then complete contour rough matching using SURF algorithm. During the SURF algorithm, construct the matching constraint of the center of the orbit waist to delete part of the false matches. Finally, the improved RANSAC algorithm is used to filter the wrong matches, achieving precise contour matching. The experimental results show that compared with traditional contour matching methods, the improved contour matching algorithm has higher accuracy and stronger robustness.