电焊机2025，Vol.55Issue(9)：98-101,144,5.DOI:10.7512/j.issn.1001-2303.2025.09.09
基于改进Mask R-CNN的焊接成形缺陷检测与区域分割研究
Research on Defect Detection and Region Segmentation of Welding Forming based on Improved Mask R-CNN
刘剑 1段瑞彬 1崔琬婷 1何亚章 1王克宽 1吴荣耀 2高嘉璘 2夏卫生2
作者信息
- 1. 中国石油集团工程技术研究有限公司,天津 300451
- 2. 华中科技大学 材料科学与工程学院 材料成形与模具技术全国重点实验室,湖北 武汉 430074
- 折叠
摘要
Abstract
To address the inefficiency of manual inspection,poor robustness of traditional machine learning methods,and the lack of precise segmentation in existing deep learning models for welding forming defects,this study proposes an im-proved Mask R-CNN-based detection model.A dataset of 638 samples(304 humps,100 craters,234 spatters)was con-structed with pixel-level mask annotations using Labelme,enhanced through data augmentation.The CBAM hybrid atten-tion module was embedded into Mask R-CNN's ResNet backbone,enabling dual channel-spatial attention mechanisms to improve sensitivity to small-target defects(e.g.,spatters).Experimental results demonstrate that the attention-enhanced model achieves 91.52%mAP@0.5(+3.34%),79.93%mAP@0.75(+9.78%),and 63.17%mAP@0.5:0.95(+4.96%)com-pared to the baseline,confirming superior effectiveness for welding defect identification and region segmentation.关键词
焊接成形缺陷/Mask R-CNN算法/缺陷识别/区域分割Key words
welding forming defects/Mask R-CNN/defect identification/regional segmentation分类
矿业与冶金引用本文复制引用
刘剑,段瑞彬,崔琬婷,何亚章,王克宽,吴荣耀,高嘉璘,夏卫生..基于改进Mask R-CNN的焊接成形缺陷检测与区域分割研究[J].电焊机,2025,55(9):98-101,144,5.
