A defect detection method for injection molding products based on multi-ray analysis

Abstract

The invention discloses a method and system for defect detection of injection molded products based on multi-ray analysis. value, obtain the corresponding grayscale curve under each light condition in different directions; obtain the position of the defect in the grayscale image under different light conditions and the depression preselection rate of the defect; obtain the depression rate of the defect in the grayscale image in the same light direction ; Determine the sag rate of the sag in the grayscale image, and judge the defect and its position in the product according to the sag rate and the grayscale curve corresponding to the sag rate. According to the grayscale information in the depression under appropriate illumination, the detection and extraction of the depression defect on the surface of the injection molded product is completed; the surface defects of the injection molded product can be detected by using the characteristics of the depression area itself, and a more accurate depression area can be obtained; no need to collect in advance Numerous images of injection molded product surfaces.

Description

technical field [0001] This application relates to the field of defect detection, in particular to a method for detecting defects of injection molded products based on multi-ray analysis. Background technique [0002] At present, the detection method of injection molding surface based on computer vision is often aimed at the relatively obvious defects in image texture features such as scratches and deformation on the injection molding surface; Segmentation is ready for detection. [0003] However, for injection molded products without circular depressions, when there are depressions on the surface of the product and there is appropriate light, low dark areas and bright areas will exist at the same time; because injection molded products have different shapes, the surface of injection molded products is not all flat , so that the low dark area and the high light area on the surface of the injection molded product are mixed together, and it is difficult to extract the depress...

AI Technical Summary

Problems solved by technology

[0003] However, for injection molded products without circular depressions, when there are depressions on the surface of the product and there is appropriate light, low dark areas and bright areas will exist at the same time; because injection molded products have different shapes, the surface of injection molded products is not all flat , so the low dark area and the bright area on the surface of the injection molded product are mixed together, and it is difficult to extract the depressed area existing in the injection molded product

[0004] The existing technology mostly uses network training for the detection of sunken areas. This method needs to collect a large number of injection molding product images that meet the standards, and does not consider the influence of light direction on the sunken area and the characteristics of the sunken area itself.

Images