4.5 Article

A strip steel surface defect detection method based on attention mechanism and multi-scale maxpooling

期刊

MEASUREMENT SCIENCE AND TECHNOLOGY
卷 32, 期 11, 页码 -

出版社

IOP Publishing Ltd
DOI: 10.1088/1361-6501/ac0ca8

关键词

attention mechanism; deep learning; defect detection; multi-scale maxpooling

资金

  1. National Key R&D Program of China [2020AAA0109300]

向作者/读者索取更多资源

The study proposes a defect detection method based on attention mechanism and multi-scale maxpooling, using Resnet50 to construct a two-stage detection model. The method was trained and tested on the NEU-DET dataset, achieving a 3.65% mAP performance improvement compared to the baseline network. Additionally, the classification accuracy of the method reaches as high as 94.73%.
In industry, defect detection involves two kinds of tasks: defect classification and location, which make it difficult to ensure the accuracy of both, and also make the task still challenging in practical application. Based on the analysis of the advantages and disadvantages of the current defect detection method, this paper proposes a defect detection method based on attention mechanism and multi-scale maxpooling (MSMP). In order to effectively improve the detection accuracy of the model, we use Resnet50 as the pre-training network construct two-stage detection model which is used to be the baseline network, and introduce the attention mechanism and MSMP module on this basis. The attention mechanism can enhance the features of the feature map extracted in each stage of Resnet50, so that the network concentrates on the effective areas for the final detection results, and ignores the background areas that are invalid or even unfavorable for detection. The proposed MSMP can incrementally enhance the receptive field, distinguish the most significant context features, and effectively improve detection precision. The proposed method is used to train and test on the NEU-DET dataset. Compared with the baseline network without any improvement, the proposed method in this paper achieves 3.65% mAP performance improvement. Meanwhile, our method achieves a performance improvement of 3.65% mAP. In addition, compared with the feature fusion mechanism, our method improves 4.03% mAP. Moreover, compared with the attention mechanisms such as spatial attention and SE block, our method improves 1.51%/1.03% mAP. Furthermore, compared with the one-stage detection algorithm SSD/YOLO-V4, the proposed method improves 5.01%/4.92% mAP. In addition, the classification accuracy of our model is as high as 94.73%.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.5
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据