Design And Implementation Of Rail Wear Detection Based On Line Laser-machine Vision

The surface and side of the rail will be worn to a certain extent during the running of the train.Railway transportation is developing in the direction of high density and high heavy load.The problem of rail wear is more prominent.How to quickly and accurately detect rail wear is an important issue at present.In this paper,based on the principle of line laser-machine vision measurement,combined with image processing and analysis technology,a rail wear detection scheme is constructed.Firstly,a variety of camera calibration methods are studied.Polynomial fitting method is used to calibrate the system.The influence of the number of calibration equations on the accuracy of camera calibration is analyzed,and the structure of the calibration equation is determined;The grid point coordinate search algorithm is given,and the camera calibration equation is solved by using the world coordinates and pixel coordinates of the grid points;Laser-covered rails are taken from different angles,image filtering processing,laser trace refinement,contour reduction and contour curve synthesis and rotation are performed on the acquired images to obtain measured rail contour data;Based on the idea of sequential similarity detection algorithm,and based on the characteristics of laser trace contour,the absolute error calculation formula is improved,and the measured contour data is registered with the standard rail data to quickly calculate the vertical wear,side wear and total wear of the current section rail.A rail wear detection device is built in the laboratory to verify the stability of the system.The experimental results show that the method can complete the rail profile detection within 100 ms,and the error is within 0.05 mm,which achieves the purpose of the expected experiment.It proves the feasibility of the method and hopes to provide reference for the development of vehicle-mounted dynamic rail wear detector.