Image color difference detection method based on feature perception and multi-channel learning

Abstract

The invention discloses an image color difference detection method based on feature perception and multi-channel learning, and the method comprises the steps: 1), constructing a training set for training a color difference detection network which is composed of a multi-channel learning module, a feature perception module, a region suggestion network and a prediction regression network; 2) inputting the image into a multi-channel learning module to obtain a comprehensive feature map of the image; 3) inputting the image comprehensive feature map into a feature sensing module to obtain a sensing weighted feature map; 4) inputting the perceptual weighted feature map into a region suggestion network to obtain a block feature map; 5) inputting the block feature map into the prediction regression network to obtain a chromatic aberration offset and a position, calculating loss with a true value, and performing back propagation to adjust parameters; 6) iteratively training to a preset value, and determining a color difference detection network; and 7) inputting the to-be-detected image into the chromatic aberration detection network to obtain chromatic aberration offset and position. According to the invention, high-speed and high-precision chromatic aberration detection of images with complex textures and patterns can be realized.

Description

technical field [0001] The invention relates to the technical field of image detection, in particular to an image color difference detection method based on feature perception and multi-channel learning. Background technique [0002] With the development of intelligence in the industrial field, more and more industrial inspection tasks begin to use visual inspection systems to replace manual inspection, such as X-ray image dangerous goods inspection, PCBA (Printed Circuit Board Assembly) defect inspection, etc. [0003] An X-ray image is an image obtained by X-ray penetrating the object to be detected and rendering different colors according to the different densities of the object. It is often used for security inspection to identify dangerous goods. PCBA is a circuit board with various components installed by patch or punching. In order to ensure the production quality, it is necessary to conduct quality inspection on the imaging of the circuit board before leaving the fac...

AI Technical Summary

Problems solved by technology

[0004] Compared with natural scene images, the above-mentioned industrial images are easily disturbed by noise or other factors during the imaging process. For example, when X-ray images are stacked or the imaging angle is distorted, the characteristics of dangerous objects are unstable, and they are easy to overlap with background features, making it difficult to Inspection: When PCBA has a complex bottom and dense components, the image formed has complex textures and patterns, and the color difference of defects is difficult to detect. If there are stains, it is more likely to miss detection and false detection

Images