Product Edge Defect Detection Method

Abstract

The application discloses a product edge defect detection method. It includes the following steps: step 1 image input; step 2 image preprocessing; step 3 template rough matching; step 4 segmentation of edge information to be tested; step 5 creating template data structure; step 6 downsampling of edge information to be tested; step 7 template Matching; step 8 calculating affine transformation parameters; step 9 template fine matching; step 10 edge defect detection. The method of the present invention adopts the edge information of the binarized image, and is insensitive to illumination changes and edge gray levels, and has the advantages of being stable and reliable; using the edge information to downsample from coarse to fine iterative template matching parameters has the advantage of fast calculation speed; using the RANSAC principle The fine matching method with parameter iterative optimization has the advantage of high detection accuracy; the template matching method using segment matching has the advantage of resistance to stretching.

Description

technical field [0001] The invention relates to a defect detection method for cover plates and plate products, in particular to a rapid stretch-resistant edge defect detection method. Background technique [0002] Industrial products often contain edges of various shapes, such as edges of circles, rectangles, and combinations of straight lines and arcs. Edge is an important feature of industrial products, which directly affects the quality of industrial products, and the defects on the edge of general products are mostly serious defects of products, such as bud missing, sawtooth, chipping and other defects directly lead to product scrapping. Product edge defect detection is an important step to ensure product quality. [0003] Edge defect detection mainly detects the consistency of the edge of industrial products, that is, detects whether there are defects such as convex marks and dents on the edge of industrial products. Specifically, for example, the edge defect detectio...

AI Technical Summary

Problems solved by technology

The research on edge defect detection algorithms at home and abroad is not very versatile. One type of algorithm uses straight line fitting to compare the distance from the edge point to the fitted line to detect edge defects in straight line areas; the other type of algorithm uses open and close operations to compare The change of the area detects the edge defects of the complete area with smooth edges; there is another type of algorithm that uses template matching to compare the similarities and differences between the test and the template to detect the edge defects of the target image

[0007] However, the above methods are not suitable for defect detection of stretched edges

Images