Weak defect enhancement multi-angle image fusion method

Abstract

The invention belongs to the related technical field of image processing, and discloses a weak defect enhancement multi-angle image fusion method, which comprises the following steps of: converting a plurality of images with different angles into weak defect images and strong defect images; constructing a background reconstruction module to extract background information in the strong defect image; constructing a defect enhancement module to extract strong defect information in the strong defect image; constructing an image fusion module, and fusing the intermediate output of the image fusion module by taking the intermediate output of the background reconstruction module and the defect enhancement module as a constraint to obtain a fusion image with an enhanced identification degree; constructing a multi-discrimination pixel-level confrontation module; constructing a comprehensive loss function, and training the comprehensive model by taking the optimal comprehensive loss function as a target to obtain a trained comprehensive model; and inputting the multi-angle image into the comprehensive model to obtain a corresponding fusion enhanced image. The defect identification degree can be improved, the problem that the defect detection precision is not high is solved, and the detection precision is greatly improved.

Description

technical field [0001] The present invention belongs to the technical field related to image processing, and more specifically relates to a multi-angle image fusion method for weak defect enhancement. Background technique [0002] In the field of industrial manufacturing, the quality of raw materials varies, and the manufacturing process is complex. Surface defects often occur on the surface of products, such as steel, wood, textiles, ceramic tiles, and new display devices such as TFT-LCD, OLED, etc. Due to the appearance of these surface defects, product quality and value will be greatly reduced, affecting user experience. [0003] At present, many algorithms have been proposed to solve the problem of surface defect detection, but these research results can only be adapted to detect surface defects with high contrast and high recognizability. For weak defects with low contrast and low recognizability, the large The defect detection methods are difficult to achieve the expe...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above defects or improvement needs of the prior art, the present invention provides a multi-angle image fusion method for weak defects enhancement, which can improve the recognizability of defects and solve the defects caused by extremely low contrast, weak brightness changes, etc. The detection accuracy is not high, and the detection accuracy is greatly improved

Images