Data enhancement method and device, data enhancement equipment and storage medium

Abstract

The invention discloses a data enhancement method, a data enhancement device, data enhancement equipment and a non-volatile computer readable storage medium. The data enhancement method comprises the following steps: acquiring a first workpiece image with defects; identifying and intercepting an image area where the defect is located in the first workpiece image to serve as a first fusion image; acquiring a defect-free second workpiece image as a second fusion image; and fusing the first fused image and the second fused image to obtain a training image. The number of the training images corresponding to the type of defects with low occurrence probability can be increased, so that the number of the training images with different types of defects is basically the same, the training effect is good when the target detection model is trained through a large number of training images subsequently, the generalization performance of the target detection model can be improved, and the defect detection effect of the target detection model is improved.

Description

technical field [0001] The present application relates to the technical field of detection, and in particular to a data enhancement method, a data enhancement device, a data enhancement device and a non-volatile computer-readable storage medium. Background technique [0002] At present, the defect of the workpiece is generally detected by the template detection algorithm, but the template detection algorithm has the problems of over-inspection and low recognition accuracy. In comparison, the neural network model with higher detection accuracy is gradually favored, but At present, during the training of the neural network model, due to the extremely uneven occurrence probability of different types of defects, the generalization performance of the model is relatively poor when using deep learning methods for wafer defect identification or detection, which directly affects the final detection accuracy. Contents of the invention [0003] The present application provides a data...

AI Technical Summary

Problems solved by technology

[0002] At present, the defect of the workpiece is generally detected by the template detection algorithm, but the template detection algorithm has the problems of over-inspection and low recognition accuracy. In comparison, the neural network model with higher detection accuracy is gradually favored, but At present, when the neural network model is trained, due to the extremely uneven occurrence probability of different types of defects, the generalization performance of the model is relatively poor when using deep learning methods for wafer defect identification or detection, which directly affects the final detection accuracy.

Images