Self-adaptive defect identification method for train wheel tread taper hole detection

Abstract

The invention discloses a self-adaptive defect identification method for train wheel tread taper hole detection, which comprises the following steps: acquiring an image to be processed, and convertinga pixel value matrix of the image from an RGB color space to a Lab color space; setting a sliding protection window and a sliding detection window, and calculating expectation and variance of pixel points in the sliding detection window, epsilon of the sliding protection window and a current detection window template A; calculating the abnormal probability of the pixel point of the current image;judging whether pixel points of the image are abnormal or not, and performing Gamma adaptive correction on the abnormal image; calculating a group of gray value variances of the corrected image; determining a threshold value by using an interpolation method, and generating a binary image; and performing feature analysis and extraction on the binarized image, screening out a suspicious area as a defect identification result, and marking the suspicious area. According to the method provided by the invention, the illumination condition can be changed in the presence of local exposure, and the damaged area can be identified and marked.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an adaptive defect recognition method for detecting tapered holes on train wheel treads. Background technique [0002] With the rapid development of high-speed railway technology, while the running speed of high-speed trains continues to increase, its safety issues should also be paid more and more attention. Train wheels are one of the key components of rail vehicles, and their health is directly related to the safety of the entire train and passengers. The friction of the wheels during emergency braking, the collision between the wheelset and the rail, and the relative sliding of the wheels to the rail surface during turning will all cause tread damage. No matter what kind of damage, it will seriously affect the quality and risk of train operation, so the image of the wheel tread It is particularly important to detect and judge whether there are defects. However, whe...

AI Technical Summary

Problems solved by technology

However, when the tread image in motion is actually taken, it is easily affected by the shooting environment and conditions. Therefore, an algorithm for the detection and recognition of tapered hole defects is studied here, and the details of the defect of the wheel tread can be displayed by using the binarization results. come out

Images