Online surface defect detection method based on photometric stereo method

Abstract

The invention relates to an online surface defect detection method based on a photometric stereo method, which comprises the following steps: in the detection process, parts continuously move along with a conveyor belt, when the parts reach a detection area, light sources distributed around the detection area are circularly lightened, and meanwhile, a camera synchronously shoots pictures under the illumination conditions of the light sources in different directions; wherein in the whole process, the parts continuously move along with a conveying belt, in the process of shooting different pictures, and the positions of the parts move; the surface defect detection step comprises the steps that a template is manufactured by extracting edge information, and unification of the positions of the parts in the multiple pictures is achieved according to template matching so as to meet the precondition of subsequent photometric stereo method calculation; then the gradient and curvature information of the surface of the part are recovered under different stations through a photometric three-dimensional method; the surface defects of the part are extracted through binarization, edge detection and connected domain detection feature extraction means in sequence, and preliminary screening is conducted according to a threshold value to judge whether the machining quality of the part to be detected is qualified or not.

Description

technical field [0001] The invention relates to a visual detection method for surface defects of parts, in particular to an online surface defect detection method. Background technique [0002] Due to the limitation of manufacturing process and transportation conditions, industrially produced parts often have surface defects during production and transportation. In order to ensure the product quality of automatic production, the surface defect detection of parts is an indispensable key link. In the past, the detection methods of such parts were often divided into manual detection and automatic detection. Among them, manual detection usually refers to human eye observation. Compared with automatic detection, manual detection is slower, reliability cannot be guaranteed, and labor costs are increased. Traditional automatic inspection methods are mostly based on machine vision technology, which collects surface images of parts and performs corresponding image processing to ext...

AI Technical Summary

Problems solved by technology

However, the traditional photometric stereo method needs to ensure that the measured part can be detected under static conditions, and it is difficult to achieve online detection

Images