Mechanical part surface defect detection method based on image processing

Abstract

The invention relates to the technical field of image processing, in particular to a mechanical part surface defect detection method based on image processing. The method comprises the following steps: acquiring a surface image of a mechanical part under a standard light source, and acquiring a segmented image of the surface image as a preliminary defect detection result; screening out a to-be-detected region in the segmented image, and performing connected domain extraction and clustering on the to-be-detected region to obtain a plurality of to-be-detected region sets; for each to-be-detected area set, moving the light source, collecting a second surface image, and obtaining feature information of the to-be-detected area set in the second surface image; obtaining an optimal light source corresponding to the maximum defect saliency according to the feature information; and obtaining an optimal image of each to-be-detected region set under the corresponding optimal light source to form a fused image, performing semantic segmentation on the fused image to obtain a second segmented image, and obtaining a detection result. According to the embodiment of the invention, the light source can be adaptively adjusted according to different defects, and an accurate defect detection result is obtained.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for detecting surface defects of mechanical parts based on image processing. Background technique [0002] During the production and processing of mechanical parts, defects will be generated on the surface of mechanical parts due to production equipment failure, raw materials mixed with sundries, and improper operation of operators. Defects have a great impact on the safe use of mechanical parts. In order to ensure the product quality, safety and life of mechanical parts in automatic production, it is an indispensable key link to detect defects on the surface of mechanical parts. [0003] Traditional automatic inspection methods are mostly based on machine vision technology, which collects surface images of parts and performs corresponding image processing to extract surface defects. However, the defect situation is often affected by the environment such as lig...

AI Technical Summary

Problems solved by technology

However, the defect situation is often affected by the environment such as lighting. It is difficult to comprehensively analyze surface defects only through image information, and sometimes even no defect information can be collected at all. The false positive rate and missed detection rate are relatively high.

Images