基于多分块均值偏移的焊缝跟踪方法、系统及存储介质
By using block processing and feature probability distribution correction, combined with least squares method and random sampling consistency, the problems of high computation time and poor anti-interference in weld seam tracking are solved, and efficient weld seam positioning and identification are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SOUTH CHINA UNIV OF TECH
- Filing Date
- 2024-01-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies fail to fully utilize prior information from images during weld seam tracking, resulting in high computational time costs and susceptibility to image interference, particularly under conditions of strong reflection and strong arc light.
A weld seam tracking method based on multi-block mean offset is adopted. The laser stripe image is processed by blocks, the feature probability distribution is calculated by kernel density estimation, the block center is iteratively corrected, and the line segment is fitted by combining the least squares method and the random sampling consistency method to obtain the weld seam location point.
It effectively reduces computation time, improves anti-interference ability, realizes accurate positioning and identification of weld seams, and expands the application range of structured light vision sensors.
Smart Images
Figure CN117893573B_ABST