Metal Surface Defect Detection Method Based on TCM-YOLO Network

doi:10.19678/j.issn.1000-3428.0069285

Abstract

Abstract:

Surface defect detection in metal production and manufacturing suffers from problems of low detection accuracy and slow processing speed. To address these problems, this study proposes a metal defect detection method based on an improved You Only Look Once version 8 (YOLOv8) network (TCM-YOLO). This method enhances the coordinate attention mechanism to the Three-Channel Coordinate Attention (TCCA) mechanism and combines it with a second version of the deformable convolutional network, i.e., the Three-channel Deformable Convolution Network (TDCN), thereby enhancing the feature extraction ability of the network. In the feature fusion network, a bidirectional feature pyramid and Dynamic Snake Convolution (DSC) are combined to improve the missed detection rate in steel strip defect detection, and to improve the retention of tiny texture and complex defect structure information. The Minimum Point Distance Intersection over Union (MPDIoU) loss function is used to replace the original loss function to accelerate the convergence speed and improve regression accuracy. Finally, a global attention mechanism is embedded to continuously capture important information regarding the global shape of the defect. Experimental results show that the average accuracy of the TCM-YOLO algorithm on the steel strip defects dataset of Northeastern University is 81.8%, which is 7.4 percentage points higher than that of the original YOLOv8 algorithm, and the accuracy reaches 78.3%, which is 8.9 percentage points higher than that of the original model. The detection speed of the algorithm reaches 61.73 frame/s. On the Tianchi aluminum profile defect dataset, the average accuracy is 4.1 percentage points higher than that of the original YOLOv8 algorithm and 8.7 percentage points higher than that of the original model. The results show that the TCM-YOLO algorithm has high detection accuracy and fast detection speed, which improves the detection capability for metal surfaces.

Key words: defect detection, object detection, YOLOv8 algorithm, attention mechanism, loss function

摘要：

针对金属生产制造中表面缺陷检测环节存在检测准确率低、处理速度慢等问题, 提出一种基于改进YOLOv8网络的金属缺陷检测方法(TCM-YOLO)。该方法改进坐标注意力机制为三通道坐标注意力机制(TCCA), 并与第2版可变形卷积网络相结合改进为三通道可变形卷积网络(TDCN), 增强网络的特征提取能力。在特征融合网络中采用双向特征金字塔与动态蛇形卷积(DSC)相结合的方法, 改善网络对于带钢缺陷检测的漏检率, 更好地保留缺陷微小纹理和复杂结构的信息。采用最小点距离交并比(MPDIoU)损失函数替换原来的损失函数加快收敛速度和获得更准确的回归结果, 最后嵌入全局注意力机制, 以不断地捕获缺陷全局形态的重要信息。实验结果表明, TCM-YOLO方法在东北大学带钢缺陷数据集上的平均精度达到了81.8%, 相比于原始的YOLOv8算法提高了7.4百分点, 精确率达到了78.3%, 相比于原模型提升了8.9百分点, 算法检测速度达到61.73帧/s, 在天池铝型材缺陷的数据集上平均精度相比于原始的YOLOv8算法提高了4.1百分点, 精确率相比于原模型提升了8.7百分点。结果表明了TCM-YOLO算法具有检测精度高、检测速度快的特点, 能更好地满足金属表面实际检测需求。

关键词: 缺陷检测, 目标检测, YOLOv8算法, 注意力机制, 损失函数

ZHAO Xiaohu, XIE Lixun, MU Dengcong, ZHANG Yue. Metal Surface Defect Detection Method Based on TCM-YOLO Network[J]. Computer Engineering, 2025, 51(6): 338-348.

赵小虎, 谢礼逊, 慕灯聪, 张悦. 基于TCM-YOLO网络的金属表面缺陷检测方法[J]. 计算机工程, 2025, 51(6): 338-348.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0069285

https://www.ecice06.com/EN/Y2025/V51/I6/338

Figures/Tables 15

Fig.1 YOLOv8 model structure

Fig.2 TDCN model structure

Fig.3 TCCA module structure

Fig.4 Different defect maps

Fig.5 GAM structure

Fig.6 Channel attention module structure

Fig.7 Spatial attention module structure

Fig.8 Dynamic snake convolution structure

Fig.9 TCM-YOLO network structure

Fig.10 Comparison of iterations of the loss function

Fig.11 Detection effects of the improved algorithms

References 26

1	杨森泉, 丁凡, 文昊翔, 等. 基于CA-YOLOv5的热轧带钢表面缺陷检测方法. 光电子·激光, 2024, 35 (1): 21- 28.
	YANG S Q , DING F , WEN H X , et al. Hot-rolled steel strip surface defects detection based on CA-YOLOv5. Journal of Optoelectronics·Laser, 2024, 35 (1): 21- 28.
2	崔克彬, 焦静颐. 基于MCB-FAH-YOLOv8的钢材表面缺陷检测算法. 图学学报, 2024, 45 (1): 112- 125.
	CUI K B , JIAO J Y . Detection algorithm of steel surface defects based on MCB-FAH-YOLOv8. Journal of Graphics, 2024, 45 (1): 112- 125.
3	贵向泉, 刘世清, 李立, 等. 基于改进YOLOv8的景区行人检测算法. 计算机工程, 2024, 50 (7): 342- 351. doi: 10.19678/j.issn.1000-3428.0068125
	GUI X Q , LIU S Q , LI L , et al. Scenic area pedestrian detection algorithm based on improved YOLOv8. Computer engineering, 2024, 50 (7): 342- 351. doi: 10.19678/j.issn.1000-3428.0068125
4	赵继达, 甄国涌, 储成群. 基于YOLOv8的无人机图像目标检测算法. 计算机工程, 2024, 50 (4): 113- 120. doi: 10.19678/j.issn.1000-3428.0068268
	ZHAO J D , ZHEN G Y , CHU F L . The UAV image target detection algorithm based on YOLOv8. Computer engineering, 2024, 50 (4): 113- 120. doi: 10.19678/j.issn.1000-3428.0068268
5	汤踊, 韩军, 魏文力, 等. 深度学习在输电线路中部件识别与缺陷检测的研究. 电子测量技术, 2018, 41 (6): 60- 65.
	TANG Y , HAN J , WEI W L , et al. Research on part recognition and defect detection of trainsmission line in deep learning. Electronic Measurement Technology, 2018, 41 (6): 60- 65.
6	李辉, 钟平, 戴玉静, 等. 基于深度学习的输电线路锈蚀检测方法的研究. 电子测量技术, 2018, 41 (22): 54- 59.
	LI H , ZHONG P , DAI Y J , et al. Study on detection method of transmission line rusty based on deep learning. Electronic Measurement Technology, 2018, 41 (22): 54- 59.
7	马燕婷, 赵红东, 阎超, 等. 改进YOLOv5网络的带钢表面缺陷检测方法. 电子测量与仪器学报, 2022, 36 (8): 150- 157.
	MA Y T , ZHAO H D , YAN C , et al. Strip steel surface defect detection method by improved YOLOv5 network. Journal of Electronic Measurement and Instrumentation, 2022, 36 (8): 150- 157.
8	程婧怡, 段先华, 朱伟. 改进YOLOv3的金属表面缺陷检测研究. 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. Computer Engineering and Applications, 2021, 57 (19): 252- 258.
9	卢俊哲, 张铖怡, 刘世鹏, 等. 面向复杂环境中带钢表面缺陷检测的轻量级DCN-YOLO. 计算机工程与应用, 2023, 59 (15): 318- 328.
	LU J Z , ZHANG C Y , LIU S P , et al. Lightweight DCN-YOLO for strip surface defect detection in complex environments. Computer Engineering and Applications, 2023, 59 (15): 318- 328.
10	冷浩, 夏骄雄. 基于改进YOLOv7的金属表面缺陷检测方法. 计算机时代, 2023 (9): 48-53, 58.
	LENG H , XIA J X . Metal surface defect detection method based on improved YOLOv7. Computer Era, 2023 (9): 48-53, 58.
11	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2016: 356-367.
12	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2023-12-20]. https://arxiv.org/abs/1804.02767.
13	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2017: 266-278.
14	吴磊, 储钰昆, 杨洪刚, 等. 面向铝合金焊缝DR图像缺陷的Sim-YOLOv8目标检测算法. 中国激光, 2024, 51 (16): 21- 30.
	WU L , CHU Y K , YANG H G , et al. Sim-YOLOv8 target detection algorithm for defects in DR images of aluminum alloy welds. Chinese Journal of Lasers, 2024, 51 (16): 21- 30.
15	油亚鹏, 马波, 赵乐, 等. 基于CA-YOLOv8的输送带大块煤检测方法[J/OL]. 计算机辅助设计与图形学学报, 2024: 1-12. [2024-02-08]. https://doi.org/10.3724/SP.J.1089.2023-00471.
	YOU Y P, MA B, ZHAO L, et al. Detection method of large coal in conveyor belt based on CA-YOLOv8[J/OL]. Journal of Computer-Aided Design& Computer Graphics, 2024: 1-12. [2024-02-08]. https://doi.org/10.3724/SP.J.1089.2023-00471. (in Chinese)
16	ZHENG Z, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[EB/OL]. [2023-12-20]. https://arxiv.org/abs/.1609.6999.
17	史涛, 崔杰, 李松. 优化改进YOLOv8实现实时无人机车辆检测的算法. 计算机工程与应用, 2024, 60 (9): 79- 89.
	SHI T , CUI J , LI S . Optimization and improvement of YOLOv8 algorithm for real-time UAV vehicle detection. Computer Engineering and Applications, 2024, 60 (9): 79- 89.
18	WANG R X, SHIVANNA R, CHENG D, et al. DCN V2: improved deep & cross network and practical lessons for web-scale learning to rank systems[C]//Proceedings of WWW'21. New York, USA: ACM Press, 2021: 1785-1797.
19	DAI J F, QI H Z, XIONG Y W, et al. Deformable convolutional networks[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2017: 764-773.
20	HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2021: 13713-13722.
21	LIU Y C, SHAO Z R, HOFFMANN N. Global attention mechanism: retain information to enhance channel-spatial interactions[EB/OL]. [2023-12-20]. https://arxiv.org/abs/2112.05561.
22	ROY K , HASAN M , RUPTY L , et al. Bi-FPNFAS: bi-directional feature pyramid network for pixel-wise face anti-spoofing by leveraging Fourier spectra. Sensors, 2021, 21 (8): 2799.
23	QI Y L, HE Y T, QI X M, et al. Dynamic snake convolution based on topological geometric constraints for tubular structure segmentation[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2023: 6070-6079.
24	曹雨淇, 徐慧英, 朱信忠, 等. 基于YOLOv8改进的打架斗殴行为识别算法: EFD-YOLO. 计算机工程与科学, 2024, 46 (10): 1825- 1834.
	CAO Y Q , XU H Y , ZHU X Z , et al. An improved fighting behavior recognition algorithm based on YOLOv8: EFD-YOLO. Computer Engineering & Science, 2024, 46 (10): 1825- 1834.
25	MA S L, XU Y, MA S L, et al. MPDIoU: a loss for efficient and accurate bounding box regression[EB/OL].[2023-12-20]. https://arxiv.org/abs/2307.07662. https://arxiv.org/abs/2307.07662v1.
26	杨阳, 杨帅, 闫敏, 等. 基于UDD-YOLO的边缘端绝缘子放电严重程度评估算法. 电子测量与仪器学报, 2024, 38 (1): 219- 227.
	YANG Y , YANG S , YAN M , et al. Evaluation algorithm of discharge severity of edge insulator based on UDD-YOLO. Journal of Electronic Measurement and Instrumentation, 2024, 38 (1): 219- 227.

[1]	HUANG Qiqiang, AN Guocheng, XIONG Gang. Open-Set Traffic Object Detection Algorithm Based on Vision-Language Pre-training Model [J]. Computer Engineering, 2025, 51(6): 375-384.
[2]	HUA Jiabao, ZHANG Jingrui, ZHU Fumin, CHEN Lu. Adaptive Spatial Transformation Method for Vehicle Detection Based on Roadside Cameras [J]. Computer Engineering, 2025, 51(6): 349-359.
[3]	FENG Xiaofei, XIE Cheng, ZHANG Xiuzhen, DONG Shikui, CHEN Junsheng, YE Shu, ZHONG Xian. Detection Method of Precast Beam Process Based on Dynamic-Static Fusion Mutual Learning [J]. Computer Engineering, 2025, 51(6): 385-394.
[4]	LIU Kai, REN Hongyi, LI Ying, JI Yi, LIU Chunping. Medical Visual Question Answering Based on Cross-Modal Attention Feature Enhancement [J]. Computer Engineering, 2025, 51(6): 49-56.
[5]	LI Yi, XU Huiying, ZHU Xinzhong, HUANG Xiao, WANG Shumeng, LI Xiyu. Mask-YOLO: Improved Mask Detection Algorithm Based on YOLOv5n [J]. Computer Engineering, 2025, 51(6): 297-310.
[6]	SHAN Pengchang, GAO Lijian, DONG Wenlong, MAO Qirong. Action Detection Method Based on Salient Target Tracking [J]. Computer Engineering, 2025, 51(6): 93-101.
[7]	XI Qi, WANG Mingjie, WEI Jinghe, ZHAO Wei. Small Object Detection Algorithm for Aerial Photography Based on Improved YOLOv3 [J]. Computer Engineering, 2025, 51(6): 184-192.
[8]	MA Yuekun, MA Mingyou. Metaphor Recognition Model Based on Weighted Integration of Global and Local Features [J]. Computer Engineering, 2025, 51(5): 143-153.
[9]	ZHOU Siyu, XU Huiying, ZHU Xinzhong, HUANG Xiao, SHENG Ke, CAO Yuqi, CHEN Chen. Mobile Phone Screen Defect Detection Algorithm Based on Improved YOLOv8n: PGS-YOLO [J]. Computer Engineering, 2025, 51(5): 326-339.
[10]	XU Huajie, ZHENG Liwen, ZHANG Pin, QIN Yuanzhuo. Lightweight Concrete Crack Detection Method Based on Multi-Dimensional Attention Module [J]. Computer Engineering, 2025, 51(5): 351-360.
[11]	HUANG Kun, QI Zhaojian, WANG Juanmin, HU Qian, HU Weichao, PI Jianyong. Aggregation Pedestrian Detection Model Based on Improved YOLOv8 [J]. Computer Engineering, 2025, 51(5): 133-142.
[12]	WANG Xiaolong, JIANG Bo, LUO Runshu, AN Guocheng. Congestion Detection Algorithm of Highway Toll Station Based on Multi-Information Fusion [J]. Computer Engineering, 2025, 51(5): 377-386.
[13]	DU Chenyang, ZHANG Xueying, HUANG Lixia, LI Juan. Multi-Feature Speech Emotion Recognition Based on Improved Efficient Channel Attention Mechanism [J]. Computer Engineering, 2025, 51(4): 97-106.
[14]	HUANG Shuoqing, HUANG Jingui. Improved Steel Defect Detection Method Based on Enhanced Fusion of RFB and YOLOv5 Features [J]. Computer Engineering, 2025, 51(4): 249-260.
[15]	DONG Hongliang, NIU Yan, SUN Yang, LI Jun. Speech Emotion Recognition Based on Memory Capsules and Attention [J]. Computer Engineering, 2025, 51(4): 169-177.

Please choose a citation manager

Content to export