Workpiece defect detection method and device fusing multi-attention mechanism

Abstract

The invention provides a workpiece defect detection method and device fusing a multi-attention mechanism. The method comprises the following steps: constructing a multi-attention defect detection model, which comprises a pyramid segmentation attention mechanism module, a channel attention mechanism module, a space self-attention mechanism module and a Unet network model; acquiring a target detection image of a to-be-detected workpiece; marking and amplifying the target detection image to obtain a secondary target detection image; dividing the secondary target detection image into a training set and a verification set; training a multi-attention deficit detection model according to the training set and the verification set; and performing defect detection on the to-be-detected workpiece by adopting the trained multi-attention defect detection model. The method has the advantages of light weight and high reasoning speed of the Unet network model, and can effectively extract multi-scale space information with finer granularity, so that a target pixel can be calculated by utilizing global information in convolution, and the segmentation precision can be improved.

Description

technical field [0001] The invention relates to the technical field of defect detection, in particular to a workpiece defect detection method integrating a multi-attention mechanism and a workpiece defect detection device integrating a multi-attention mechanism. Background technique [0002] In the production process of industrial products, quality inspection is a key link, and the quality inspection of appearance defects on the product surface is a very common problem in the manufacturing industry. Industrial products need to inspect the surface state of the product before leaving the factory. It is necessary to obtain the edge contour information of the defect and the pixel resolution of the defect to evaluate the level of the defect, which is convenient for the quality inspector to customize the severity of the inspection and optimize the production process of the production workshop. . The difference between many defects on the surface of the workpiece and the backgroun...

AI Technical Summary

Problems solved by technology

The difference between many defects on the workpiece surface and the background is very small, and the gray scale gradually transitions, which brings difficulties to the defect segmentation with blurred boundaries

[0003] In addition, traditional defect detection algorithms generally require a large amount of defect detection data, but the amount of defect detection data on the workpiece surface is extremely limited, which makes it difficult to meet the requirements of model training data volume, resulting in insufficient segmentation accuracy of the model and difficult to meet the requirements of actual production

Images