A domain adaptation YOLOv5 model for industrial defect inspection. (31st May 2023)

• Record Type: Journal Article
• Authors: Li, Chen; Yan, Haoxin; Qian, Xiang; Zhu, Shidong; Zhu, Peiyuang; Liao, Chengwei; Tian, Haoyang; Li, Xiu; Wang, Xiaohao; Li, Xinghui
• Is Part Of: Measurement. Volume 213(2023)
• Publication Date: 2023-05-31
• Subjects: Industrial inspection -- Deep learning -- Transfer learning -- Surface defect -- Magnetic tiles; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2023.112725
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: Propose a new DAVOLOv5 model based on domain adaptation. Add a transfer term to the loss function. Use model ensemble methods for hard-to-detect samples. Recommend a hyperparameter controlling the domain representations. Abstract: In the process of industrial production, manufactured products are prone to surface defects for a variety of reasons. To overcome the problem of high time cost and the strong demand for large sample data sets, a detector based on transfer learning is commonly utilized. In this paper, a domain adaptation YOLOv5 model, named DAYOLOv5, is proposed for automatic surface defect inspection. The hyperparameter α in DAYOLOv5 for knowledge transfer can be designed specially to achieve better generalization in real-world industrial applications. Meanwhile, in the field of magnetic tile surface defect detection, our DAYOLOv5 outperforms traditional mixed training and pretrain-finetune methods with limited data sets and has great robustness. Overall, the experimental results demonstrate that our DAYOLOv5 model can indeed boost performance on small-scale target data sets and is applicable to practical industrial scenarios.