中国机械工程2024，Vol.35Issue(9)：1634-1641,8.DOI:10.3969/j.issn.1004-132X.2024.09.013
基于距离感知的金属缺陷样本标签分配算法
Label Assignment Algorithm for Metal Defect Samples Based on Distance-awareness
摘要
Abstract
The distance-aware dynamic label assignment(DDA)algorithm was proposed to address issues such as the lack of consideration for aspect ratio of metal surface defects and poor localization ability towards target distribution during the allocation of positive and negative samples in training processes of metal surface defect detection model.DDA did not change the structures of original detec-tion model and did not increase computational expenses.A new distance loss calculation paradigm was proposed based on geometric characteristics of real frame to optimize the regression problem with a wide aspect ratio.The regression offset in iterative processes was decoded as predicted frame coordi-nates.Finally,the comprehensive intersection and union ratio information were calculated among the predicted frame,anchor frame and real frame,and positive and negative samples were dynamically se-lected to improve training accuracy.It was verified through the surface defect detection task of cold-rolled strip in a steel plant in Wuhan,and a public hot-rolled strip surface defect data set was intro-duced for generalization testing.The detection results are significantly improved,which has practical application values for metal surface quality specifications.关键词
目标检测/样本选择策略/宽高比/金属缺陷检测/距离回归损失函数Key words
object detection/sample selection strategy/aspect ratio/metal defect detection/dis-tance regression loss function分类
信息技术与安全科学引用本文复制引用
朱传军,梁泽启,付强,张超勇..基于距离感知的金属缺陷样本标签分配算法[J].中国机械工程,2024,35(9):1634-1641,8.基金项目
国家自然科学基金国际(地区)合作与交流项目(51861165202) （地区）
广东省重点领域研发计划(2019B090921001) （2019B090921001）
