Zipper production defect intelligent detection method and system based on optical means

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent zipper production defect detection method and system based on optical means, and the system is an artificial intelligence system in the field of zipper production and can realize information system integration service of the artificial intelligence system. The method comprises the following steps: acquiring the unevenness difference of each pixel point; according to the abnormal degree, the tooth width and the standard tooth width of each uneven chain tooth, the weight of each uneven chain tooth is obtained; performing spatial sine curve fitting on the uneven area, solving curve parameters by using fitting errors of starting points and ending points of a plurality of chain teeth in the uneven area and weights of corresponding pixel points to obtain a fitting curve, and calculating actual tooth pitches between adjacent chain teeth according to the fitting curve; and performing defect detection on the zipper according to the tooth length, the tooth width and the actual tooth pitch. According to the embodiment of the invention, the defect of the zipper is tested or analyzed by using an optical means, particularly a visible light image, so that the defect detection accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an intelligent detection method and system for zipper production defects based on optical means. Background technique [0002] In the production process of metal zippers, there are often defects in the size of the chain teeth and the uneven tooth pitch. Since the position of each sprocket relative to the light source is different, the glossiness and brightness of the sprockets in the captured images are different, and the zipper may not be flat. Due to the camera imaging, the sprockets or tooth pitches in the image are different from the actual ones. [0003] At present, the method of defect detection for zipper is to directly analyze the photographed image and compare it with the standard zipper image to detect the defect position. However, the photographed image may be affected by the light, which makes the image of the chain teeth different from the actual one,...

AI Technical Summary

Problems solved by technology

[0002] In the production process of metal zippers, there are often defects of different sizes and uneven tooth pitches.

Because the position of each sprocket to the light source is different, the glossiness of the sprocket in the captured image is different, and the zipper may be uneven. Due to the camera imaging, the sprocket or tooth pitch in the image is different from the actu...

Images