Three-dimensional point cloud-based surface defect detection method and device and storage medium

Abstract

The invention discloses a three-dimensional point cloud-based surface defect detection method, which comprises the following steps: acquiring an original point cloud, and scanning the surface of a sample to be detected by a three-dimensional camera to obtain the original point cloud; obtaining the contribution degree of each point in the original point cloud through a simplified sampling algorithm, and forming a to-be-tested point cloud by the points with the contribution degrees greater than or equal to a preset contribution degree in the original point cloud; obtaining a standard model point cloud, and registering the to-be-measured point cloud and the standard model point cloud, so that the to-be-measured point cloud and the standard model point cloud are aligned to the same position; and comparing the registered to-be-detected point cloud with the standard model point cloud to detect the defect, and obtaining the defect characteristic quantity of the surface of the to-be-detected sample. According to the invention, the original point cloud is subjected to simplified sampling, coarse registration and fine registration, so that the calculation speed of the point cloud data can be increased, and the calculation efficiency is improved. And through a three-position surface defect detection mode of point-to-reconstruction surface projection, the defect information of the surface of the sample can be accurately detected on the premise of not influencing the precision.

Description

technical field [0001] The present invention relates to the technical field of machine vision and image processing, in particular to a surface defect detection method, device and storage medium based on a three-dimensional point cloud. Background technique [0002] Surface defects of products mainly include differences in volume, shape, length, etc., such as protrusions, depressions, cavities, and damage on the surface of products. Defects in the appearance of the product not only affect the user experience, but also may cause losses to users, and even cause safety accidents. For example, defects on the surface of train rails will reduce their service life, posing safety hazards. [0003] At present, the defect detection on the product surface is mainly divided into manual detection and automatic detection. Among them, manual inspection relies on human eyes to detect product appearance quality, but human eyes will be tired, and different people have different judgment stand...

AI Technical Summary

Problems solved by technology

However, the original point cloud obtained by using a 3D camera has a large order of magnitude, and the distribution is complex and uneven, and there is a lot of noise and redundant data, which reduces the efficiency and accuracy of defect detection; the 3D point cloud data is discrete and difficult to process , the original two-dimensional image processing algorithm is no longer applicable

Images