Method of defect detection, device and equipment

Abstract

The invention relates to a method of defect detection, device and equipment. The method comprises steps: in the case of detecting whether a surface defect exists in an electronic product included in aspecified image, a first well-trained neural network model is used to classify the specified image, wherein the first neural network model is used for detecting whether the image belongs to a problematic image or a normal image; if the first neural network model classifies the specified image to the problematic image and the given reliability value of the specific image belonging to the problematic image is smaller than a reliability threshold, a second well-trained neural network model is used to classify the specified image, wherein the second neural network model is used for detecting whether the image belongs to a problematic image or a normal image; and if the second neural network model classifies the specified image to the problematic image, the surface defect is determined to exist in the electronic product included in the specified image. The method, the device and the equipment can improve the defect detection accuracy in a condition of consuming no manpower basically.

Description

technical field [0001] The invention relates to the field of deep learning, in particular to a method, device and equipment for defect detection. Background technique [0002] On the production line of electronic products such as chips or computer processors (CPU), it is necessary to detect whether there are surface defects in the produced electronic products, for example, whether there are stains, hairline cracks and / or white spots on the electronic products. [0003] The current inspection process is to pass the produced electronic products through the inspection equipment in turn, and then the inspection equipment takes multiple images of each passing electronic product from different angles, and selects from the images taken for each electronic product The electronic products contained therein are very likely to have suspected images of surface defects, and finally the inspectors visually inspect each of the screened suspected images to determine whether the electronic p...

AI Technical Summary

Problems solved by technology

[0004] In the case of a large output of electronic products, the number of suspect images screened by the detection equipment will also be large. In this case, it will be very labor-intensive for the inspectors to detect a large number of suspect images

Moreover, inspectors are prone to fatigue after working for a period of time, so there is a possibility of false detection

Therefore, the overall defect detection efficiency detected by inspectors is not high, and the detection accuracy is not stable enough

Images