改进基于YOLOv8的钢材表面缺陷检测算法
Improvement of the Steel Surface Defect Detection Algorithm Based on YOLOv8
DOI: 10.12677/csa.2025.155126, PDF,    科研立项经费支持
作者: 孔灏铮：江苏理工学院机械工程学院，江苏 常州
关键词: 钢材表面；缺陷检测；YOLOv8；GAM注意力机制；P2检测头；Steel Surface； Defect Detection； YOLOv8； GAM Attention Mechanism； P2 Detection Head
摘要: 为提高钢材表面缺陷小目标检测效率，提出一种改进基于YOLOv8的钢材表面缺陷检测算法。通过添加GAM注意力机制，它通过结合了通道注意力和空间注意力，来增强图像特征的表示能力，有助于提升图像处理任务的性能。同时，为了提高小目标检测准确性和更好的多尺度信息融合，特此在里面引入了P2小目标检测头。在NEU-DET数据集上做消融和对比实验，改进算法与原算法比较，precision提高了2.5个百分点；recall提高了3.3个百分点；mAP@0.5提高了2.1个百分点；mAP@0.5：0.95提高了3.5个百分点。
Abstract: To enhance the efficiency of small target detection for surface defects on steel, we propose an improved steel surface defect detection algorithm based on YOLOv8. This algorithm incorporates the GAM attention mechanism, which integrates both channel and spatial attention, thereby enhancing the representation capability of image features and contributing to improved performance in image processing tasks. Additionally, to further increase the accuracy of small target detection and facilitate better multi-scale information fusion, we introduce a P2 small target detection layer. We conduct ablation and comparative experiments on the NEU-DET dataset. Compared to the original algorithm, the improved algorithm demonstrates an increase in precision by 2.5 percentage points, recall by 3.3 percentage points, mAP@0.5 by 2.1 percentage points, and mAP@0.5:0.95 by 3.5 percentage points.
文章引用：孔灏铮. 改进基于YOLOv8的钢材表面缺陷检测算法[J]. 计算机科学与应用, 2025, 15(5): 539-547. https://doi.org/10.12677/csa.2025.155126

参考文献

[1] Jian, C., Gao, J. and Ao, Y. (2017) Automatic Surface Defect Detection for Mobile Phone Screen Glass Based on Machine Vision. Applied Soft Computing, 52, 348-358. [Google Scholar] [CrossRef] 
[2] He, Z. and Liu, Q. (2020) Deep Regression Neural Network for Industrial Surface Defect Detection. IEEE Access, 8, 35583-35591. [Google Scholar] [CrossRef] 
[3] Liu, Y., Xiao, H., Xu, J. and Zhao, J. (2022) A Rail Surface Defect Detection Method Based on Pyramid Feature and Lightweight Convolutional Neural Network. IEEE Transactions on Instrumentation and Measurement, 71, 1-10. [Google Scholar] [CrossRef] 
[4] Singh, S.A. and Desai, K.A. (2022) Automated Surface Defect Detection Framework Using Machine Vision and Convolutional Neural Networks. Journal of Intelligent Manufacturing, 34, 1995-2011.
[5] Liang, F., Zhou, Y., Chen, X., Liu, F., Zhang, C. and Wu, X. (2021) Review of Target Detection Technology Based on Deep Learning. Proceedings of the 5th International Conference on Control Engineering and Artificial Intelligence, Sanya, 14-16 January 2021, 132-135. [Google Scholar] [CrossRef] 
[6] Girshick, R. (2015) Fast R-CNN. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 7-13 December 2015, 1440-1448. [Google Scholar] [CrossRef] 
[7] Zhang, K. and Shen, H. (2021) Solder Joint Defect Detection in the Connectors Using Improved Faster-RCNN Algorithm. Applied Sciences, 11, Article 576. [Google Scholar] [CrossRef] 
[8] Yang, A., Jiang, T., Han, Y., Li, J., Li, Y. and Liu, C. (2022) Research on Application of On-Line Melting In-Situ Visual Inspection of Iron Ore Powder Based on Faster R-CNN. Alexandria Engineering Journal, 61, 8963-8971. [Google Scholar] [CrossRef] 
[9] Kumar, A. and Manikandan, R. (2021) Brain Tumor Detection Using Deep Neural Network-Based Classifier. In: Khanna, A., Gupta, D., Bhattacharyya, S., Hassanien, A.E., Anand, S. and Jaiswal, A., Eds., International Conference on Innovative Computing and Communications, Springer, 173-181. [Google Scholar] [CrossRef] 
[10] Guo, F., Qian, Y., Rizos, D., Suo, Z. and Chen, X. (2021) Automatic Rail Surface Defects Inspection Based on Mask R-CNN. Transportation Research Record: Journal of the Transportation Research Board, 2675, 655-668. [Google Scholar] [CrossRef] 
[11] Ren, J.S. and Wang, Y. (2022) Overview of Object Detection Algorithms Using Convolutional Neural Networks. Journal of Computer and Communications, 10, 115-132.
[12] Jiang, P., Ergu, D., Liu, F., Cai, Y. and Ma, B. (2022) A Review of Yolo Algorithm Developments. Procedia Computer Science, 199, 1066-1073. [Google Scholar] [CrossRef] 
[13] Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A. and Zagoruyko, S. (2020) End-to-End Object Detection with Transformers. In: Vedaldi, A., Bischof, H., Brox, T. and Frahm, J.M., Eds., Computer Vision—ECCV 2020, Springer, 213-229. [Google Scholar] [CrossRef] 
[14] 蔡剑锋, 柏俊杰, 张雪等. 基于改进Mask R-CNN的金属板材表面缺陷检测[J]. 重庆科技学院学报(自然科学版), 2023, 25(2): 110-116.
[15] Zhou, S., Zeng, Y., Li, S., Zhu, H., Liu, X. and Zhang, X. (2022) Surface Defect Detection of Rolled Steel Based on Lightweight Model. Applied Sciences, 12, Article 8905. [Google Scholar] [CrossRef] 
[16] Yang, L., Huang, X., Ren, Y. and Huang, Y. (2022) Steel Plate Surface Defect Detection Based on Dataset Enhancement and Lightweight Convolution Neural Network. Machines, 10, Article 523. [Google Scholar] [CrossRef] 
[17] 张政超. 改进YOLOv5的轻量级带钢表面缺陷检测[J]. 计算机系统应用, 2023, 32(6): 278-285.
[18] Qin, R., Chen, N. and Huang, Y. (2022) EDDNet: An Efficient and Accurate Defect Detection Network for the Industrial Edge Environment. 2022 IEEE 22nd International Conference on Software Quality, Reliability and Security (QRS), Guangzhou, 5-9 December 2022, 854-863. [Google Scholar] [CrossRef] 
[19] 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.