Product edge defect detection method

Abstract

The invention relates to a product edge defect detection method, which comprises the following steps that firstly, a template image and a sample image are input, and then a feature map is constructed by acquiring information such as a centroid, a target contour, a centroid and a deflection angle from the template image and the sample image; an iterative optimization method of a mapping model from a sample image to a template image is constructed by taking a minimum residual sum as a loss function, so that feature map matching is carried out, a global mapping matrix is obtained according to a matching result, a difference image process is completed, and a coarse segmentation region is obtained by using an adaptive threshold value; secondly, for the problem of scarcity of industrial image samples, a large number of artificial defect sample pre-training data sets are obtained through modeling defects, the sample sets are used for pre-training the multi-scale integrated residual neural network, and then a real defect sample set is used for migration training; the obtained result is used for carrying out defect type identification on the coarse segmentation region.

Description

technical field [0001] The invention belongs to the technical field of glass detection, and relates to a glass surface defect detection method, in particular to a defect detection method of a glass cover plate of a mobile phone. Background technique [0002] Over the past few years, with the popularization of 5G in my country, the increase in the speed of wired networks and the decline in tariffs, more and more network users have shifted from PC to mobile. In 2020, the proportion of mobile Internet users will reach 99.2%, so the number and growth rate of mobile phones are increasing year by year. Because the glass cover has better mechanical and optical properties than metal, and the cost is lower than ceramics, it is favored by many mobile phone companies. However, in the process of actual production, manufacturing and transportation, various types of damage to the glass cover will inevitably occur. Therefore, timely and strict quality inspection in the production process...

AI Technical Summary

Problems solved by technology

Among them, there is an algorithm that directly predicts the location and type of defects through neural networks. Its disadvantage is that it relies heavily on manually labeled data sets. However, due to the particularity and scarcity of industrial images, it is difficult to obtain large-scale, same-type defect samples. It also directly affects its detection accuracy and effect

The other is to use the traditional image processing method, which uses the method of directly extracting the defect feature vector and then classifying it. The disadvantage is that it relies heavily on the adjustment of artificial algorithms and has poor stability.

Images