A defect target detection method based on an attention mechanism

Abstract

The invention belongs to the technical field of defect detection, and discloses a defect target detection method based on an attention mechanism, which comprises the following steps: marking various defects of all pictures in an original data set to obtain a standard training data set with marks; Obtaining a training label according to the standard training data set, determining a loss function, obtaining a network model, and performing training by using a reverse conduction method to obtain a defect regression detection network model based on an attention mechanism and having enhanced defectpart weight; Performing classification prediction and regression prediction on the to-be-detected picture by utilizing the defect regression detection network model; Carrying out non-maximum suppression processing on the predicted defect bounding box and filtering the defect bounding box to obtain an output result; According to the method provided by the invention, the weight of the defect area isimproved through an attention mechanism, so that the defect detection precision is improved; The industrial product surface defect classification and regression detection method can be applied to other types of surface defect detection frameworks to improve the detection precision, and is high in universality.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and more specifically, relates to a defect target detection method based on an attention mechanism. Background technique [0002] Object detection includes two tasks: object category prediction (that is, classification) and object bounding box regression. These two tasks share the features of the candidate frame extracted by the convolutional neural network, and have achieved good results in object detection in natural scenes. However, it has not been widely used in industrial inspection tasks to detect product surface defects. Compared with general object detection, the surface defect detection of industrial products is often not like natural scene object detection, which has many categories and complex backgrounds. Although the characteristics of defects can be learned by using conventional detection methods, the proportion of defects relative to the original image is very high. Small,...

AI Technical Summary

Problems solved by technology

However, when it is applied to industrial inspection tasks, the technology of detecting product surface defects has not been widely used.

Compared with general object detection, the surface defect detection of industrial products is often not like natural scene object detection, which has many categories and complex backgrounds. Although the characteristics of defects can be learned by using conventional detection methods, the proportion of defects relative to the original image is very large. Small, and the direction of appearance is arbitrary, the range of aspect ratio is large, the range of scale is large, and the labeling is not so accurate

On the other hand, due to the relatively low probability of defects during product production, it is difficult to collect data sets; and for industrial products, there is a high requirement for missed detection rates, which makes the detection of industrial product surface defects a relatively large challenging

Images