Wang Han, Yang Xiuding, Zhou Bei, Shi Zhuohao, Zhan Daohua, Huang Renbin, Lin Jian, Wu Zhiheng, Long Danfeng
School of Mechanical and Electrical Engineering, Guangdong University of Technology, Guangzhou 510006, China.
Institute of Intelligent Manufacturing, Guangdong Academy of Sciences, Guangzhou 510070, China.
Materials (Basel). 2023 Mar 31;16(7):2811. doi: 10.3390/ma16072811.
In order to improve the detection accuracy of the surface defect detection of industrial hot rolled strip steel, the advanced technology of deep learning is applied to the surface defect detection of strip steel. In this paper, we propose a framework for strip surface defect detection based on a convolutional neural network (CNN). In particular, we propose a novel multi-scale feature fusion module (ATPF) for integrating multi-scale features and adaptively assigning weights to each feature. This module can extract semantic information at different scales more fully. At the same time, based on this module, we build a deep learning network, CG-Net, that is suitable for strip surface defect detection. The test results showed that it achieved an average accuracy of 75.9 percent (mAP50) in 6.5 giga floating-point operation (GFLOPs) and 105 frames per second (FPS). The detection accuracy improved by 6.3% over the baseline YOLOv5s. Compared with YOLOv5s, the reference quantity and calculation amount were reduced by 67% and 59.5%, respectively. At the same time, we also verify that our model exhibits good generalization performance on the NEU-CLS dataset.
为提高工业热轧带钢表面缺陷检测的精度，将深度学习的先进技术应用于带钢表面缺陷检测。本文提出一种基于卷积神经网络（CNN）的带钢表面缺陷检测框架。具体而言，提出一种新颖的多尺度特征融合模块（ATPF），用于融合多尺度特征并为每个特征自适应分配权重。该模块能更充分地提取不同尺度的语义信息。同时，基于此模块构建了适用于带钢表面缺陷检测的深度学习网络CG-Net。测试结果表明，它在6.5千兆浮点运算（GFLOPs）和每秒105帧（FPS）的情况下，平均精度达到75.9%（mAP50）。检测精度比基线YOLOv5s提高了6.3%。与YOLOv5s相比，参数量和计算量分别减少了67%和59.5%。同时，我们还验证了我们的模型在NEU-CLS数据集上表现出良好的泛化性能。
