Magnetic core surface defect detection method based on deep learning

Abstract

The invention discloses a magnetic core surface defect detection method based on deep learning, which comprises the following steps of: firstly, acquiring all front and back images of a magnetic core, preprocessing, extracting defect parts, inputting a deep convolutional generative adversarial network improved by a Gaussian mixture model, training an extracted defect set to generate a new defect image, performing poisson fusion on the obtained new image and the intact magnetic core image, making a standard data set, and dividing a training set and a verification set; inputting the training set image into a YOLO-v3 neural network for training, proposing a new training strategy, setting training parameters, and taking the trained network as a magnetic core surface defect detection model so as to identify category information and position information of the magnetic core. The method provided by the invention is faster and more accurate in identification.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a method for detecting surface defects of magnetic cores based on deep learning. Background technique [0002] With the continuous development of society, the degree of industrial automation is getting higher and higher. However, while improving production efficiency, the problem of product defects still plagues many manufacturers. Moreover, products with quality problems flow into the market, which may cause greater economic losses and safety threats. At present, most manufacturers still use manual quality inspection methods to detect product defects, which greatly wastes human resources, and there are still cases of missed inspections and false inspections. In order to improve the problems existing between human beings and speed up the detection time and detection accuracy, it has become the trend of industrial development to replace manual detection with machine vision, pattern...

AI Technical Summary

Problems solved by technology

[0004] In view of the low quality of existing magnetic core defect detection, the current situation of missed detection and false detection, the present invention proposes a detection method for magnetic core surface defects based on deep learning. The specific technical scheme is as follows:

Images