Micro-Defect Detection Algorithm with Improved Attention Mechanism in Section Steel Surface

doi:10.3778/j.issn.1002-8331.2110-0375

Abstract

Abstract: A novel surface defect detection method named DMSA-YOLOv3 based on dual multi-scale attention mechanism is proposed in order to solve the problems of low efficiency and poor accuracy of surface defects detection caused by various and numerous micro-scale defects, which achieves a rapid and accurate detection of section steel surfaces. Firstly, a dual multi-scale attention, i. e. DMSA model is constructed based on channel and space attention to screen the fuse different scale features for strengthening the weight of small-scale defect features. Secondly, the feature extraction backbone network of YOLOv3 model is processed by deep separable convolution to reduce the size of model for speeding the detection procedure, and a multi-scale long-distance context feature extraction layer in order to improve the ability of extracting small defect targets’ feature of backbone network is built. Finally, a DMSA-YOLOv3 defect detection model combined DMSA model and improved YOLOv3 model is constructed and applied to multi-scale defect detection of steel surface. The experimental results show that DMSA-YOLOV3 model has a multi-category average detection accuracy of 97.6% and a detection speed of 55.3?frame/s, which is 4.7 percentage points and 24.5?frame/s higher than YOLOv3 model respectively. The minimum size defects of 10×10（mm2） can be detected, which is 6.25 times higher than that of YOLOv3 model. The DMSA-YOLOv3 model has boosted the accuracy and speed of surface defects detection of steel prominently.

Key words: section steel, defect detection, attention mechanism, YOLOv3 model

摘要： 针对型钢表面缺陷种类多样、微小缺陷占比较大导致的检测效率低、检测精度差的问题，提出了一种基于双重多尺度注意力机制的表面缺陷检测方法DMSA-YOLOv3，实现型钢表面多尺度缺陷快速精确检测。构建了基于通道和空间的双重多尺度注意力模型DMSA，对不同尺度特征进行筛选融合，强化小尺度缺陷的特征权重；改进了YOLOv3模型，使用深度可分离卷积对DarkNet53特征提取主干网络实现轻量化处理，提高检测速度，并构建多尺度长距离上下文特征提取层，使用4种不同扩张率的并行空洞卷积替代全局池化，提高模型对小尺寸缺陷的特征提取能力；构建了融合DMSA模型和改进YOLOv3模型的DMSA-YOLOv3缺陷检测模型，并应用于型钢表面多尺度缺陷检测。实验结果表明：DMSA-YOLOv3模型具有97.6%的多类别平均检测精度和55.3?frame/s的检测速度，与YOLOv3模型相比分别提升了4.7个百分点和24.5?frame/s；最小可检出20×20像素（约10×10?mm2）缺陷，与YOLOv3模型相比提高了6.25倍，有效提升了型钢表面缺陷的检测精度与检测速度。

关键词: 型钢, 缺陷检测, 注意力机制, YOLOv3模型

YU Haitao, LI Fulong, LIU Yajiao, WANG Jiang, YU Lifeng, ZHANG Chunhui, LIU Baoshun, MA Yongfu. Micro-Defect Detection Algorithm with Improved Attention Mechanism in Section Steel Surface[J]. Computer Engineering and Applications, 2022, 58(11): 250-259.

于海涛, 李福龙, 刘亚姣, 王江, 于利峰, 张春晖, 刘宝顺, 马永福. 改进注意力机制的型钢表面微小缺陷检测方法[J]. 计算机工程与应用, 2022, 58(11): 250-259.

References

[1] IHOR K，PAVLO M，JANETTE B，et al.Steel surface defect classification using deep residual neural network[J].Metals-Open Access Metallurgy Journal，2020，10（6）：846-861.
[2] 李少波，杨静，王铮，等.缺陷检测技术的发展与应用研究综述[J].自动化学报，2020，46（11）：2319-2336.
LI S B，YANG J，WANG Z，et al.Review of development and application of defect detection technology[J].Acta Automatica Sinica，2020，46（11）：2319-2336.
[3] 李国和，乔英汉，吴卫江，等.深度学习及其在计算机视觉领域中的应用[J].计算机应用研究，2019，36（12）：3521-3529.
LI G H，QIAO Y H，WU W J，et al.Review of deep learning and its application in computer vision[J].Application Research of Computers，2019，36（12）：3521-3529.
[4] 胡晓辉，张建国.基于改进卷积神经网络的图像超分辨率算法研究[J].计算机应用研究，2020，37（3）：947-950.
HU X H，ZHANG J G.Research on image super-resolution algorithm based on improved convolutional neural network[J].Application Research of Computers，2020，37（3）：947-950.
[5] 陶显，侯伟，徐德，等.基于深度学习的表面缺陷检测方法综述[J].自动化学报，2020，47（5）：1017-1034.
TAO X，HOU W，XU D，et al.A survey of surface defect detection methods based on deep learning[J].Acta Automatica Sinica，2020，47（5）：1017-1034.
[6] GIRSHICK R，DONAHUE J，DARRELL T，et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2014.
[7] GIRSHICK R.Fast-RCNN[C]//International Conference on Computer Vision（ICCV），2015.
[8] REN S，HE K，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Learning，2017，39（6）：1137-1149.
[9] HE K，GKIOXARI G，DOLLAR P，et al.Mask R-CNN[C]//IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2018.
[10] 王立中，管声启.基于深度学习算法的带钢表面缺陷识别[J].西安工程大学学报，2017，31（5）：669-674.
WANG L Z，GUAN S Q.Strip steel surface detect recognition based on deep learning[J].Journal of Xi’an Polytechnic University，2017，31（5）：669-674.
[11] 翁玉尚，肖金球，夏禹，等.改进Mask R-CNN算法的带钢表面缺陷检测[J].计算机工程与应用，2021，57（19）：235-242.
WENG Y S，XIAO J Q，XIA Y，et al.Strip surface defect detection based on improved Mask R-CNN algorithm[J].Computer Engineering and Applications，2021，57（19）：235-242.
[12] 韩强，张喆，续欣莹，等.特征融合和级联检测网络的Faster R-CNN钢材表面缺陷检测算法[J].太原理工大学学报，2021，62（1）：127-135.
HAN Q，ZHANG Z，XU X Y，et al.Faster R-CNN steel surface defect detection algorithm based on feature fusion and cascade detection network[J].Journal of Taiyuan University of Technology，2021，62（1）：127-135.
[13] 王兵，乐红霞，李文璟，等.改进YOLO轻量化网络的口罩检测算法[J].计算机工程与应用，2021，57（8）：62-69.
WANG B，YUE H X，LI W J，et al.Mask detection algorithm based on improved YOLO lightweight network[J].Computer Engineering and Applications，2021，57（8）：62-69.
[14] REDMON J，DIVVALA S，GIRISHICK R，et al.You only look once：unified，real-time object detection[C]//Computer Vision & Pattern Recognition，Las Vegas，June 26-July 1，2016：779-788.
[15] 程婧怡，段先华，朱伟，等.改进YOLOv3的金属表面缺陷检测研究[J].计算机工程与应用，2021，57（19）：252-258.
CHENG J Y，DUAN X H，ZHU W，et al.Study on metal surface defect detection by improved YOLOv3[J].Application Research of Computers，2021，57（19）：252-258.
[16] 袁野，谭晓阳.复杂环境下的冰箱金属表面缺陷检测[J].计算机应用，2021，41（1）：270-274.
YUAN Y，TAN X Y.Defect detection of refrigerator metal surface in complex environment[J].Journal of Computer Applications，2021，41（1）：270-274.
[17] 吴湘宁，贺鹏，邓中港，等.一种基于注意力机制的小目标检测深度学习模型[J].计算机工程与科学，2021，43（1）：95-104.
WU X N，HE P，DENG Z G，et al.A deep learning model of small object detection based on attention mechanism[J].Computer Engineering and Science，2021，43（1）：95-104.
[18] HUA X，WANG X，RUI T，et al.A fast self-attention cascaded network for object detection in large scene remote sensing images[J].Applied Soft Computing，2020，94（6）：106-117.
[19] LIANG X，ZHANG J，ZHUO L，et al.Small object detection in unmanned aerial vehicle images using feature fusion and scaling-based single shot detector with spatial context analysis[J].IEEE Transactions on Circuits and Systems for Video Technology，2020，30（6）：1758-1770.