Design Of Rail Profile Detection System Online Based On Machine Vision

The steady development of China's economy continues to drive the development of rail transit,high-speed passenger transport and heavy-duty freight transport have become the main directions for China's medium and long-term railway network planning.With the development trend of high-speed railways and heavy-duty railways,the quality of rail products has received extensive attention.In order to improve the quality assurance capability of rail products,many advanced detection technologies and equipment are continuously applied to major rail production plants,among them,machine vision technology has the advantage of accurately and quickly detecting rail external defects and has gradually become popular in the rail manufacturing industry.In order to solve the problems of poor manual detection,low efficiency,and large deviation in Rail and Beam Plant of Pangang,this paper obtains the full-profile contour size of the rail through the on-line detection system based on machine vision technology to avoid manual detection of risk and data deviation.Based on the rail cross-section geometry,the measurement system needs to measure the various dimensions,determine the detection range that the detection system must reach,formulate measurement schemes for each inspection size,propose the functional requirements of the detection system,and form the actual requirements for rail profile detection,as a basic condition for carrying out system design work.By analyzing the principle of structured light triangulation,the control structure of the rail profile detection system is established,and the key parts and components of the equipment are selected according to the detection requirements to meet the actual working conditions at the production site.The on-line profile detection system of rails is tested on the line,and the corresponding system test data are obtained according to the test method requirements,the key and measurable dimensions are selected for manual measurement,the data obtained by the two methods are analyzed and compared,aim at the problem which the data's accurate and comprehensive,identify the corresponding optimization solutions,test again,and verify the optimization results.