Research On Track Crawling Displacement Measurement System Based On Image Processing

After long-term operation and use of the track,the phenomenon of track crawling is prone to occur.The measurement of the crawling displacement can effectively prevent the occurrence of the track crawling phenomenon and ensure the development of the national economy and the safety of life.The development of image processing technology has realized the non-contact measurement of track crawling displacement.All the railways in operation in my country have many monitoring points that need to measure the creeping displacement.However,the existing non-contact measurement methods cannot automatically distinguish different monitoring points of the track,and it is difficult to count the displacement data after the measurement is completed.Therefore,this paper aims to study a tracking displacement measurement system based on image processing,which has the advantages of a high degree of automation.The main research work of the paper is as follows.（1）A non-contact track displacement measurement scheme is proposed.First,a coding marker is designed to identify the coding value of the marker,which can automatically distinguish different monitoring points of the track;then analyze different displacement quantification methods,and select a binocular measurement method with low equipment cost and high precision to realize the marker Conversion of pixel coordinates to space coordinates;finally,the camera parameters and camera calibration principles are analyzed,and a dual-target calibration program based on OpenCV is written.（2）Research on the image recognition and processing algorithm of markers.The main research contents are the method of locating the center of the marker,the method of matching the center of the circle,and the method of coding recognition.According to the multiconstraint information provided by the marker,an optimal positioning method of the center of the circle is studied to obtain the coordinates of the center of the marker with sub-pixel accuracy.A method of circle center matching based on epipolar geometry is studied to reduce the incorrect matching points of the circle center,and the purpose is to restore the spatial circle center coordinates with high precision.A coding recognition method is designed,which mainly performs one-dimensional pixel scanning and processing on the coding region of markers.（3）Design the hardware and software of the track displacement measurement system.First,the hardware circuit and application program of the image acquisition side are designed to realize the timing acquisition and transmission of images;then,based on the research content of marker image recognition and processing,combined with the QT development framework,the image processing side is designed to provide a friendly interactive interface for the staff.（4）Build a track displacement measurement system,and carry out displacement measurement experiments,comparative measurement experiments,and code recognition experiments.In the displacement measurement experiment,the system measurement error was calculated by simulating the occurrence of track crawling to verify the feasibility of the system;in the comparative measurement experiment,the influence of distance change and angle change on the measurement accuracy of the system was studied.In the code recognition experiment,the accuracy of marker code recognition under different measurement distances was studied.The final experimental results show that within a distance of 2000 mm,the measurement error of the system is ±0.5mm,and the code recognition accuracy is 100%,indicating that the displacement measurement system designed in this paper is feasible in actual track engineering.