Research And Implementation Of Rail Head Full Profile Wear Detection System Based On Laser Vision Technology

In the daily operation of urban rail transit,the track plays an important role as a carrier for train travel.The quality of the track condition directly determines the safety of the train operation: on the one hand,the train is easily affected by the track bending condition during actual travel,and the existence of a small radius curve at the track bend will always cause huge centrifugal force when the train turns Greatly increased the wear of the wheelset treads facing the inside of the rail.On the other hand,due to the large number of passengers riding on the subway every day,the track is used too frequently,causing wear on its surface and inside.This wear condition will not only increase the running resistance of the train,but also reduce the comfort and safety of the ride.Therefore,the detection of track wear has become the daily maintenance task of the railway.Compared with foreign advanced track inspection technology,at present,China still conducts manual inspections mainly on rail wear measurement rulers.This detection method is not only time-consuming and laborious,but also the accuracy of the detection and the safety of the inspection workers cannot be guaranteed.Therefore,in view of the difficulties and pain points of the current rail wear detection,the development of high-efficiency,high-precision,low-cost rail wear inspection system has become the focus of development.In view of the above problems,this article has carried out the research and development of the rail wear inspection system based on the Shenzhen metro rail wear detection requirements.The main contents are:(1)Categorize and elaborate the existing or under-developed rail wear detection methods at home and abroad,and propose the detection methods in this paper according to the needs of subway detection and the experience of workers’ inspection.(2)Complete the test plan design and hardware system construction for the entire detection system,including camera,lens,line laser selection and image acquisition platform design and construction.(3)Through the analysis of the camera perspective projection model,study and establish the transformation relationship of the four coordinate systems.By comparing and analyzing the camera calibration methods,the Zhang-type calibration method is finally selected to design and complete the camera calibration experiment,and the calibration errors are quantitatively analyzed and displayed.(4)First of all,the rail contour image collection and pre-processing is completed,and the fusion algorithm based on the variable direction template method and the gray center of gravity method is adopted to achieve the extraction of the center line of the light bar in combination with the line laser contour characteristics.Secondly,according to the profile characteristics of the left and right rail heads of the rail,the overall contour reconstruction of the rail head is completed.Finally,according to the characteristics of the reconstructed rail head contour and the ICP algorithm,the contour is accurately matched and the vertical wear,side wear and total wear values are calculated.(5)Complete the integrated joint debugging of the entire measurement system,complete the rail wear test on the Shenzhen Metro Line 2,and analyze the results for errors.The accuracy of the measurement system and the feasibility of the scheme are verified by comparing the measurement data of the rail wear inspection caliper.