Railway train wheel set surface defect detection method based on deep learning

Abstract

The invention discloses a deep learning-based railway train wheel set surface defect detection method. The method comprises the following steps of (1) acquiring a train wheel set surface image through a train wheel set surface defect detection system; (2) marking the acquired surface images, expanding train wheel set images in a data enhancement mode, and dividing an enhanced image data set into a training set, a verification set and a test set; (3) establishing a train wheel pair surface defect detection model based on deep learning, inputting the training set and the verification set into the train wheel pair surface defect detection model in batches for training, and obtaining a trained train wheel pair surface defect detection model; and (4) inputting the tested data set into the trained train wheel set surface defect detection model to obtain the shape and position of the defect in the test set image. According to the method, the defect feature information in the sample is extracted by learning the sample image, so that missing detection of fine defects in the defect image is avoided, and the detection precision and the detection speed are improved.

Description

technical field [0001] The invention relates to a method for detecting surface defects of train wheels, in particular to a method for detecting surface defects of railway train wheels based on deep learning. Background technique [0002] Railway transportation is the core link in the operation chain of the logistics industry. With the rapid development of railway transportation, stricter requirements have been put forward for the management and maintenance of train wheelsets. The working environment of the train wheelset is harsh and accompanied by long-term operation, cracks, scratches, dents and other defects often appear on the surface of the train wheelset. Therefore, the detection link for the surface defects of the train wheelset is to maintain the railway locomotive. An important guarantee for safe operation. [0003] There are two traditional methods for detecting surface defects of train wheels: artificial visual inspection method and traditional machine vision in...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: Aiming at the shortcomings of the deep learning defect detection model in the prior art for the poor recognition accuracy of small target defects, and the detection model needs to save a large number of parameters in order to learn deep features, the present invention provides a railway train based on deep learning Wheelset surface defect detection method. The train wheelset surface defect detection model established by this detection method is based on the MobileNetV2 network and U-shaped architecture. The feature extraction ability of the surface defect detection model for subtle defects is enhanced through the dense connection of multi-level features. The model utilizes different levels of features, and reduces the number of model parameters, reducing the amount of computation

Images