机械与电子2026，Vol.44Issue(1)：65-71,7.
基于YOLOv11的焊缝识别跟踪算法研究
Research on Welding Seam Identification and Tracking Algorithm Based on YOLOv11
王高伟 1杨飞 1杨成林 1刘亚东 1胡明明 2刘新华2
作者信息
- 1. 国能神东煤炭集团有限责任公司寸草塔煤矿,内蒙古 鄂尔多斯 017209
- 2. 中国矿业大学机电工程学院,安徽 徐州 221116
- 折叠
摘要
Abstract
To address the problems with weld seam detection accuracy and real-time tracking ability in the field of automated welding,a weld seam identification and tracking method based on YOLOv11 is proposed.The method analyzes the visual features of weld seams,constructs a dataset suitable for weld seam detection,and integrates the laser stripe central line extraction algorithm with the YOLOv11 model,thereby achieving high-precision detection and real-time tracking of weld seams during the welding process.The experimental results show that,at a resolution of 512×512,the proposed method achieves an identification accuracy of 99.51%,with a maximum error of 1.05 pixels in weld seam feature point recogni-tion,an inference speed of 54 ms per image,and an average location error of less than 2.2 mm.These re-sults demonstrate that the proposed method can effectively meet the strict demands of weld seam quality monitoring in intelligent manufacturing production lines.关键词
焊缝识别/焊缝跟踪/特征提取/深度学习Key words
weld seam identification/weld seam tracking/feature extraction/deep learning分类
矿业与冶金引用本文复制引用
王高伟,杨飞,杨成林,刘亚东,胡明明,刘新华..基于YOLOv11的焊缝识别跟踪算法研究[J].机械与电子,2026,44(1):65-71,7.
