Fastener positioning method and device and readable storage medium

Abstract

The embodiment of the invention provides a fastener positioning method and device and a readable storage medium, and the method comprises the steps: carrying out the preprocessing of an obtained imagecomprising a to-be-positioned fastener, and obtaining a corresponding edge region image; wherein a plurality of connected regions contained in the edge region image form a first region set; removingdominant noise and dominant interference regions in the first region set based on the shape features of each region in the first region set and the edge region features of the steel rails and the sleepers to form a second region set; based on the shape feature of each region in the second region set and the relative position feature between the regions, obtaining the linear position of the edge ofthe steel rail; based on the shape feature of each region in the second region set, the relative position feature between the regions and the steel rail edge linear position, obtaining the sleeper edge linear position; and obtaining a fastener position based on the steel rail edge linear position and the sleeper edge linear position. The method is high in positioning precision, simple in processing process, small in calculation amount and suitable for detection environments of various speeds.

Description

technical field [0001] The invention relates to the technical field of fastener positioning, in particular to a fastener positioning method, device and readable storage medium. Background technique [0002] Fasteners are very common and important connectors on railway tracks. With the continuous increase of train speed, frequency of use and load capacity, the failure rate of fasteners also increases rapidly. Therefore, the health status detection of fasteners is one of the important contents of railway track detection. Along with the rapid development of computer technology and CMOS (Complementary Metal Oxide Semiconductor, Complementary Metal Oxide Semiconductor) technology, fastener detection based on image processing has become a new trend, and the first step of this method is to detect the fasteners in the image. Components are positioned. [0003] The most widely used fastener positioning method is the cross positioning method, that is, first determine the position of...

AI Technical Summary

Problems solved by technology

This method has high precision, but due to the use of wavelet transform and gray-scale statistical map, the process is complicated and the amount of calculation is too large

[0004] In addition, the above-mentioned methods are all analyzed in the way of grayscale statistics, the statistical information contains all noise and interference information, and no noise and interference are eliminated

But for the actual orbit, the existence of objects such as gravel and branches will generate a lot of noise or interference information, which will affect the accuracy of the algorithm

Images