面向接触轨几何参数检测的插值法和图像边缘检测研究

打开文本图片集

中图分类号：TP391 文献标志码：B doi：10.20213/j.cnki.tdcl.2025.01.22.01

Abstract： In the rail transit field，the accurate measurement of the geometric parameters of contact rail is a key factor to guarantee the safety and eficiency of train operation. To satisfy the need of high-precision detection，the article studies the interpolation method and the image edge detection method to increase the accuracy and reliability of the geometric parameter detection of contact rail，adopts the image pretreatment method to increase the accuracy of interpolation and uses the improved Harrs corner-point detection and Sobel edge detection operator to extract key characteristics.According to the test results，the interpolation method has higher accuracy in sub-pixel corner-point extraction，with an absolute average error of 0.0398% ，which is better than 0.189 1% of the conventional grayscale gradient method，while the Sobel edge detection operator performs well in handling light stripe scatering noise.In practical applications，the detection model built in this article is similar to the manually detected data，with an error within 1mm . The geometric parameter detection model of contact rail built in this article significantly increases the accuracy and efficiency of the railway track detection and alows to prevent potential track faults， providing a strong technical support for the

stable operation of the railway transport system.

Key words： image pretreatment; contact rail; geometric parameter detection; interpolation method; image edge detection; sub-pixel accuracy

随着城市轨道交通的快速发展，对接触轨几何参数的精确测量需求日益迫切。（剩余8319字）