Solder spot defect detection method and system based on image processing

By collecting point cloud data from different perspectives and registering and combining, the local chromaticity difference value and normal vector continuity are calculated using the group-optimal intelligent algorithm, the problem of distinguishing between weld areas and defect areas is solved, and efficient welding joint defect detection is achieved.

CN120259321BActive Publication Date: 2025-08-12JIANGSU EASY WASH INTELLIGENT TECHNOLOGY CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202510759967.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-06-09
Publication Date
2025-08-12
Estimated Expiration
2045-06-09

AI Technical Summary

Technical Problem

During the welding process, the weld area is difficult to distinguish from the defect area, resulting in the weld area being misidentified as a defect area, affecting the service life and stability of the washing machine.

Method used

By collecting point cloud data from different perspectives, registering and merging, the final point cloud data is obtained, and the group-optimal intelligent algorithm is used to calculate the local chromatic difference value and normal vector continuity of the optimal radius, and identifying defect areas with clustered eigenvalues.

Benefits of technology

Effectively distinguishing between weld areas and defect areas, improving the accuracy and stability of defect detection, avoiding noise interference, and improving the reliability of welding joint defect detection.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention relates to the field of image processing technology, and in particular to a method and system for detecting solder joint defects based on image processing. The method comprises the steps of: collecting point cloud data from different viewing angles, registering and merging the point cloud data from different viewing angles to obtain final point cloud data, obtaining local chromaticity difference values and local density values of a point within a neighborhood of any radius, constructing an objective function based on the local chromaticity difference values and local density values, and then obtaining an optimal radius, and then obtaining the local chromaticity difference value of each point within the neighborhood of the optimal radius, obtaining the normal vector continuity of all points within the optimal radius, and obtaining a clustering feature value of each point based on the normal vector continuity and the local chromaticity difference value; clustering all points based on the clustering feature values, and then identifying defect clusters. The defect clusters identified by the present invention are more accurate.
Need to check novelty before this filing date? Find Prior Art