Method for detecting multiple scratches on metal surface containing screws

Abstract

The invention belongs to the field of metal surface defect detection, and designs a method for detecting multiple scratches on a metal surface containing screws. According to the method, the metal surface is denoised based on a user-defined filtering algorithm; and then a user-defined screw removal algorithm is used for removing the metal surface screw, then scratches are searched to improve the detection robustness and precision, and under the condition of complex surface texture interference, tiny scratches on the metal surface and even scratches covering the screw can still be detected. According to the designed metal surface scratch detection method containing the screw, interference caused by complex reflective textures of the metal surface is overcome, meanwhile, the influence of additional noise of the screw and the like is eliminated, and the high scratch detection rate is achieved.

Description

Technical field [0001] The invention relates to a method for detecting multiple scratches on a metal surface containing screws. More specifically, the invention relates to a method that can overcome screw interference and realize simultaneous detection of multiple scratches. Background technique [0002] Among the various surface defects of mechanical products, scratch defects are the most common. Due to the influence of the processing technology and even the transportation operation, the scratch itself has the characteristics of small scale and large shape difference. However, the scratch detection process is very susceptible to the influence of light and interference from the background, and feature extraction is often difficult. Therefore, the detection of scratch defects has always been a difficult point in product surface quality detection. In order to accurately detect scratches and make up for the low efficiency of manual inspection and different inspection standards, mac...

AI Technical Summary

Problems solved by technology

However, in the detection of metal surface scratches by machine vision, the texture of the metal surface is a great interference factor. After being photographed by an industrial camera and binarized, the texture will be displayed as scratches. Therefore, there are many problems in the extraction and detection of scratches. high difficulty

In order to solve the problem of detecting scratches on metal surfaces or even detecting multiple scratches at the same time, the present invention designs a scratch detection method applied to metal surfaces containing screws based on image morphology, custom filtering algorithm and custom screw removal algorithm. method

Images