锻压技术

基于RM-YOLOv8的热轧带钢表面缺陷检测

英文标题：Surface defect detection on hot rolled strip steel based on RM-YOLOv8

分类号：TP391.4

出版年,卷(期):页码：2025,50(2):103-114

摘要：

针对传统热轧带钢表面缺陷检测算法在复杂工业环境下检测精度低、易漏检和误检等问题，提出一种基于RM-YOLOv8的热轧带钢表面缺陷检测算法。在特征提取网络中引入ConvNeXt Block模块，增强了网络特征提取能力；构建了注意力机制RC_MHSA，加强了特征提取网络对于特征图中感兴趣区域的训练；构建了轻量可学习视觉中心模块LLVC，并以跨层通道融合的方式向深层网络中加入更加丰富的特征信息，提高了模型对于局部角落特征的捕捉；最后，利用GhostBottleneck模块构造了一种轻量化模块C2f_Ghost，在减少计算量的同时，提高了网络的检测精度。实验结果表明，RM-YOLOv8在热轧带钢表面缺陷数据集NEU-DET上实验的平均精度均值相比于原始YOLOv8网络提升了7.0%，精确率提高了8.9%，召回率提高了6.8%。与其他检测算法相比，提出的RM-YOLOv8缺陷检测算法具有更好的检测性能和工业应用价值。

For the problems of low detection accuracy, easy to miss detection and misdetect in complex industrial environments caused by traditional algorithms for the surface defect detection on hot rolled strip steel, a surface defect detection algorithm of hot rolled strip steel based on RM-YOLOv8 was proposed, and ConvNeXt Block module was introduced into the feature extraction network to enhance the feature extraction capability of the network. Then, attention mechanism RC_MHSA was constructed to enhance the training of feature extraction network for the regions of interest in feature maps, and a lightweight and learnable visual center module (LLVC) was constructed. Furthermore, the richer feature information was added to the deep network by cross-layer channel fusion to improve the model ability to capture local corner features, and finally a lightweight module C2f_Ghost was constructed by using GhostBottleneck module, which reduced the amount of computation and improved the detection accuracy of the network. The experimental results show that the average mean precision of RM-YOLOv8 on the NEU-DET dataset of surface defects in hot rolled strip steel is increased by 7.0%, the accuracy is increased by 8.9%, and the recall rate is increased by 6.8% compared to the original YOLOv8 network. Compared with other detection algorithms, the proposed RM-YOLOv8 defect detection algorithm has better detection performance and industrial application value.

基金项目：

陕西省自然科学基础研究计划项目（2023-JC-YB-452）

作者简介：杨澳(1999-)，男，硕士研究生,E-mail：2075330899@qq.com;通信作者：侯红玲(1976-)，女，博士，教授,E-mail：xjtuhhl@163.com

参考文献：

[1]Liu G H, Ma Q W. Strip steel surface defect detecting method combined with a multi-layer attention mechanism network[J]. Measurement Science and Technology, 2023, 34(5):055403.

[2]Chen W, Zou B, Huang C Z, et al. The defect detection of 3D-printed ceramic curved surface parts with low contrast based on deep learning[J]. Ceramics International, 2023, 49(3):2881-2893.

[3]Li W H, Zhang H O, Wang G L, et al. Deep learning based online metallic surface defect detection method for wire and arc additive manufacturing[J]. Robotics and Computer-Integrated Manufacturing, 2023, 80:102470.

[4]Lu L, Hou J, Yuan S Q, et al. Deep learning-assisted real-time defect detection and closed-loop adjustment for additive manufacturing of continuous fiber-reinforced polymer composites [J]. Robotics and Computer-Integrated Manufacturing, 2023, 79:102431.

[5]Wang L. Application of deep learning to detect defects on the surface of steel balls in an IoT environment [J]. The Journal of Supercomputing, 2022, 78(14):16425-16452.

[6]王泽.无缝钢管直径测量与表面缺陷检测系统研究[D].桂林:广西师范大学,2019.

Wang Z. Research on Seamless Steel Tubes Diameter Measurement and Surface Defect Detection System[D]. Guilin: Guangxi Normal University, 2019.

[7]石磊,汪建余,孙胜伟,等.基于视觉检测技术的冲压收料线监测系统开发[J].锻压技术,2023,48(9):184-189.

Shi L, Wang J Y, Sun S W, et al， Development on monitoring system for stamping receiving line based on visual inspection technology [J]. Forging & Stamping Technology, 2023, 48 (9): 184-189.

[8]阎馨,杨月川,屠乃威.基于改进SSD的钢材表面缺陷检测[J].现代制造工程,2023(5):112-120.

Yan X, Yang Y C, Tu N W. Steel surface defect detection based on improved SSD [J]. Modern Manufacturing Engineering, 2023(5): 112-120.

[9]Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector [A]. European Conference on Computer Vision[C]. Cham：Springer, 2016.

[10]张亚腾,黄俊.基于YOLOv7的钢表面缺陷检测[J].激光杂志,2024,45(3):87-93.

Zhang Y T, Huang J. Steel surface defect detection based on YOLOv7 [J]. Laser Journal, 2024, 45(3): 87-93.

[11]Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[A]. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) [C]. Vancouver, 2023.

[12]寇旭鹏,刘帅君,麻之润.基于Faster-RCNN的钢带缺陷检测方法[J].中国冶金,2021,31(4):77-83.

Kou X P, Liu S J, Ma Z R. Defect detection method of steel strip based on Faster-RCNN [J]. China Metallurgy, 2021,31(4): 77-83.

[13]Ren S Q, He K M, Girshick R. Faster R-CNN: Towards real-time object detection with region proposal networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.

[14]代小红,陈华江,朱超平.一种基于改进Faster RCNN的金属材料工件表面缺陷检测与实现研究[J].表面技术,2020,49(10):362-371.

Dai X H, Chen H J, Zhu C P. Surface defect detection and realization of metal workpiece based on improved Faster RCNN [J]. Surface Technology, 2020, 49(10): 362-371.

[15]Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[A]. 2014 IEEE Conference on Computer Vision and Pattern Recognition[C]. Columbus, 2014.

[16]Liu Z, Mao H, Wu C Y, et al. A ConvNet for the 2020s[DB/OL]. https://arxiv.org/abs/2201.03545.

[17]Bi J, Hare J S, Merrett G V. GhostShiftAddNet: More features from energy-efficient operations[DB/OL]. https://arxiv.org/abs/2109.09495.

[18]Ge Z, Liu S T, Wang F, et al. YOLOX: Exceeding YOLO series in 2021[DB/OL]. https://arxiv.org/abs/2107.08430.

[19]Li C, Li L, Jiang H, et al. YOLOv6: A single-stage object detection framework for industrial applications[DB/OL]. https://arxiv.org/abs/2209.02976.

[20]Lin T Y, Dollar P, Girshick R, et al. Feature pyramid networks for object detection[A]. Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition[C]. Honolulu: IEEE, 2017.

[21]孙泽强,陈炳才,崔晓博,等.融合频域注意力机制和解耦头的YOLOv5带钢表面缺陷检测[J].计算机应用,2023, 43(1):242-249.

Sun Z Q, Chen B C, Cui X B, et al. Strip steel surface defect detection by YOLOv5 algorithm fusing frequency domain attention mechanism and decoupled head [J]. Journal of Computer Applications, 2023, 43(1): 242-249.

[22]Howard A G, Zhu M L, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications[DB/OL]. https://arxiv.org/abs/1704.04861.

[23]Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16×16 words: Transformers for image recognition at scale[A]. Proceedings of the 2021 International Conference on Learning Representations[C]. Virtual Event: OpenReview.net, 2021.

[24]Wortsman M, Lee J, Gilmer J, et al. Replacing softmax with ReLU in vision transformers [DB/OL]. http://arxiv.org/abs/2309.08586.

[25]Quan Y, Zhang D, Zhang L Y, et al. Centralized feature pyramid for object detection[J]. IEEE Trans. on Image Process., 2023, 32: 4341-4354.

[26]Woo S, Park J, Lee J Y, et al. CBAM: Convolutional block attention module[A]. Proceedings of the 2018 European Conference on Computer Vision（ECCV）[C]. Cham: Springer, 2018.

[27]Hu J, Shen L, Sun G. Squeeze-and-excitation networks [A]. Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition[C]. Salt Lake City: IEEE, 2018.

[28]Hou Q B, Zhou D Q, Feng J S. Coordinate attention for efficient mobile network design[A]. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition[C]. Los Alamitos: IEEE Computer Society Press, 2021.

[29]GitHub-Inc. YOLOv5 [EB/OL]. https://github.com/ultralytics/yolov5, 2021-06-15.

[30]程婧怡,段先华,朱伟.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. Research on metal surface defect detection by improved YOLOv3 [J]. Computer Engineering and Applications, 2021, 57(19): 252-258.

[31]徐洪俊,唐自强,张锦东,等.钢材表面缺陷检测的YOLOv5s算法优化研究[J].计算机工程与应用,2024,60(7):306-314.

Xu H J, Tang Z Q, Zhang J D, et al. Research on optimization of YOLOv5s detection algorithm for steel surface defect [J]. Computer Engineering and Applications, 2024, 60(7): 306-314.

[32]李少雄,史再峰,孔凡宁,等.一种面向钢材表面缺陷检测的改进型YOLOv5算法[J].激光与光电子学进展,2023,60(24):184-192.

Li S X, Shi Z F, Kong F N, et al. An improved YOLOv5 algorithm for steel surface defect detection [J]. Laser & Optoelectronics Progress, 2023, 60(24): 184-192.

[33]高春艳,秦燊,李满宏,等.改进YOLOv7算法的钢材表面缺陷检测研究[J].计算机工程与应用,2024,60(7):282-291.

Gao C Y, Qin S, Li M H, et al.Research on steel surface defect detection based on improved YOLOv7 algorithm[J]. Computer Engineering and Applications, 2024, 60(7): 282-291.

服务与反馈：

【文章下载】【加入收藏】