Screen defect detection method, device and system, computer equipment and medium

Abstract

The invention discloses a screen defect detection method, device and system, computer equipment and a medium. A specific embodiment of the method comprises the steps of inputting a to-be-detected screen image into an enhanced learning network to generate an intermediate image; inputting a target training set formed by the defect-free screen image and the intermediate image into a discriminator ofthe generative adversarial network to generate a discrimination result, feeding back the discrimination result to the reinforcement learning network as a return value of the reinforcement learning network until the discrimination result meets a preset convergence condition, thereby obtaining a background reconstruction image; and differentiating the to-be-detected screen image and the background reconstruction image to obtain a defect image. According to the embodiment, background reconstruction of the to-be-detected screen image can be performed based on the mutually constrained reinforcementlearning network and the generative adversarial network so that the defect image capable of clearly presenting the defect position and the defect quantization level can be obtained, and visual and accurate defect detection can be realized.

Description

technical field [0001] The invention relates to the technical field of image processing. More specifically, it relates to a screen defect detection method, device and system, computer equipment and media. Background technique [0002] The screen defects of common display screens are divided into point defects, line defects and mura defects. Among them, the types of point defects are complex and diverse, the geometric shape is not fixed, and the size, brightness or grayscale are close to noise. Therefore, general preprocessing of images with point defects is very likely to "blur" the characteristics of point defects. Even remove it as noise; mura defects are specially used to represent a type of surface defect with low contrast, uneven brightness, and an area larger than one pixel. It is the most complex and difficult type of defect to detect, so the realization of defect images Effective segmentation is very difficult. [0003] At present, the display screen, especially t...

AI Technical Summary

Problems solved by technology

Among them, the types of point defects are complex and diverse, the geometric shape is not fixed, and the size, brightness or grayscale are close to noise. Therefore, general preprocessing of images with point defects is very likely to "blur" the characteristics of point defects. Even remove it as noise; mura defects are specially used to represent a type of surface defect with low contrast, uneven brightness, and an area larger than one pixel. It is the most complicated and difficult type of defect to detect, so the defect image It is very difficult to effectively segment

[0003] At present, the display screen, especially the display module of the VR device, has higher and higher requirements for PPI and higher requirements for the process. The traditional manual screen defect detection method has difficulties, low detection efficiency, and missed detection. High efficiency, and great waste of materials and production capacity

However, the existing more advanced methods of screen defect detection through machine vision are susceptible to the interference of regular background textures, so there will still be problems of false detection, missed detection and inaccurate defect positioning.

Images