基于多尺度特征增强与融合的冷轧带钢缺陷检测方法研究

doi:10.3901/JME.260145

摘要/Abstract

摘要： 冷轧带钢中表面缺陷的高效检测对产品质量控制至关重要，然而缺陷常呈现多尺度特征，且常受到复杂背景严重干扰。基于此，提出一种轻量级的检测框架—EAF-YOLO。该方法引入高效空间坐标注意力模块，提升模型对复杂背景下缺陷区域的感知能力；设计多尺度特征提取模块，增强对不同尺度缺陷的表征能力；构建自适应注意力卷积模块，实现多尺度特征的动态融合与增强。试验结果表明EAF-YOLO在保持较低计算复杂度的同时，平均精度(mAP50)达到88.0%，实现了高精度与低计算开销的平衡。现场部署结果充分验证了EAF-YOLO 模型在实际生产环境中的适用性，各类缺陷的分类精度均达到 90% 以上。

关键词: 表面缺陷检测, 卷积神经网络, 多尺度特征提取, 注意力机制

Abstract: Detecting surface defects in cold-rolled steel strips efficiently is crucial for quality control. However, such defects often exhibit multi-scale characteristics and are frequently obscured by complex backgrounds. To address this challenge, this paper proposes a lightweight detection framework—EAF-YOLO. The method incorporates an Efficient Spatial Coordinate Attention Module (ESCAM) to enhance the model's ability to perceive defect regions under complex backgrounds; a Multi-scale Feature Extraction module (MFE) to improve the representation of defects at different scales; and an Adaptive Attention Convolution module (AAConv) to enable dynamic fusion and enhancement of multi-scale features. Experimental findings demonstrate that EAF-YOLO maintains low computational complexity while attaining an average accuracy (mAP50) of 88.0%, striking a balance between high accuracy and minimal computational overhead. Results from on-site deployment thoroughly validate the practicality of the EAF-YOLO model in real production settings, with classification accuracy for various defects exceeding 90%.

Key words: surface defect detection, convolutional neural network, multi-scale feature extraction, attention mechanism

中图分类号:

陈树宗, 付天添, 蒋圣泉, 华长春. 基于多尺度特征增强与融合的冷轧带钢缺陷检测方法研究[J]. 机械工程学报, 2026, 62(12): 202-215.

CHEN Shuzong, FU Tiantian, JIANG Shengquan, HUA Changchun. Research on Cold-rolled Strip Steel Defect Detection Method Based on Multi-scale Feature Enhancement and Fusion[J]. Journal of Mechanical Engineering, 2026, 62(12): 202-215.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260145

http://www.cjmenet.com.cn/CN/Y2026/V62/I12/202

参考文献

[1] 张殿华，丁成砚，王云龙，等. 板带轧制数字化技术进步与发展趋势[J]. 轧钢，2024，41(5):51-65. ZHANG Dianhua，DING Chengyan，WANG Yunlong，et al. Progress and development trends of digital technology in strip rolling[J]. Rolling，2024，41(5):51-65.
[2] 孙杰，张殿华. 板带材轧制过程数字化模型与智能优化[C]//中国金属学会，中国金属学会青年工作委员会. 第十二届中国金属学会青年学术年会暨首届“碳中和”冶金青年科学家沙龙论文集(一)，2024. SUN Jie，ZHANG Dianhua. Digital model and intelligent optimization of strip rolling process[C]// Chinese Society for Metals，Youth Committee of Chinese Society for Metals. Proceedings of the 12th Annual Academic Conference of the Chinese Society for Metals and the First “Carbon Neutrality” Metallurgical Young Scientists Salon (I)，2024.
[3] 李旭，金树仁，王鹏飞，等. 冷轧板形控制技术的研究现状及展望[J]. 中国冶金，2024，34(7):21-30. LI Xu，JIN Shuren，WANG Pengfei，et al. Research status and prospects of cold rolling shape control technology[J]. China Metallurgy，2024，34(7):21-30.
[4] 郑建聪，谢麒麟，方挺，等. 一种基于特征点的管材表面缺陷视觉检测方法[J]. 安徽工业大学学报(自然科学版)，2022，39(1):21-24. ZHENG Jiancong，XIE Qilin，FANG Ting，et al. A feature point-based visual inspection method for surface defects on pipes [J]. Journal of Anhui University of Technology (Natural Science Edition)，2022，39(1):21-24.
[5] 司念文，张文林，屈丹，等. 卷积神经网络表征可视化研究综述[J]. 自动化学报，2022，48(8):1890-1920. SI Nianwen，ZHANG Wenlin，QU Dan，et al. A survey on convolutional neural network representation visualization[J]. Acta Automatica Sinica，2022，48(8):1890-1920.
[6] 何彧，宋克臣，张德富，等. 融合多层级特征的弱监督钢板表面缺陷检测算法[J]. 东北大学学报(自然科学版)，2021，42(5):687. HE Yu，SONG Kechen，ZHANG Defu，et al. A weakly supervised steel surface defect detection algorithm based on multi-level feature fusion[J]. Journal of Northeastern University (Natural Science Edition)，2021，42(5):687.
[7] 曾治霖，瞿昊，杜正春. 基于深度学习和生成对抗网络的发动机缸体表面缺陷检测方法[J]. 机械工程学报，2025，61(2):46-55. ZENG Zhilin，QU Hao，DU Zhengchun. Surface defect detection method for engine block based on deep learning and generative adversarial networks[J]. Journal of Mechanical Engineering，2025，61(2):46-55.
[8] ZHAO S，LI G，ZHOU M，et al. ICA-Net:Industrial defect detection network based on convolutional attention guidance and aggregation of multiscale features[J]. Engineering Applications of Artificial Intelligence，2023，126:107134.
[9] ZHANG L，CHEN J，CHEN J，et al. LDD-Net:Lightweight printed circuit board defect detection network fusing multi-scale features[J]. Engineering Applications of Artificial Intelligence，2024，129:107628.
[10] WAN F，ZHANG G，LI Z. DFSDNet:A dual-branch multi-scale feature fusion network for surface defect detection of copper strips and plates[J]. Computers in Industry，2025，167:104265.
[11] YU X，LYU W，WANG C，et al. Progressive refined redistribution pyramid network for defect detection in complex scenarios[J]. Knowledge-Based Systems，2023，260:110176.
[12] ZHANG E，MA Q，CHEN Y，et al. EGD-Net:Edge-guided and differential attention network for surface defect detection[J]. Journal of Industrial Information Integration，2022，30:100403
[13] CHEN Y，DING Y，ZHAO F，et al. Surface defect detection methods for industrial products:A review[J]. Applied Sciences，2021，11(16):7657.
[14] ZHANG Z，WANG X，JUNG C. DCSR:Dilated convolutions for single image super-resolution[J]. IEEE Transactions on Image Processing，2018，28(4):1625-1635.
[15] 张宸嘉，朱磊，俞璐. 卷积神经网络中的注意力机制综述[J]. 计算机工程与应用，2021，57(20):64-72. ZHANG Chenjia，ZHU Lei，YU Lu. A survey of attention mechanisms in convolutional neural networks[J]. Journal of Computer Engineering and Applications，2021，57(20):64-72.
[16] XUE F，PANG L，HUANG X，et al. A lightweight convolutional network based on RepLKnet and ghost convolution[C]//2024 4th International Conference on Electronic Information Engineering and Computer Communication (EIECC). IEEE，2024:1571-1576.
[17] 李文涛，彭力. 多尺度通道注意力融合网络的小目标检测算法[J]. 计算机科学与探索，2021，15(12):2390-2400. LI Wentao，PENG Li. A small object detection algorithm based on multi-scale channel attention fusion network[J]. Journal of Frontiers of Computer Science and Technology，2021，15(12):2390-2400.
[18] CHATTERJEE S，KON E，SHARMA P，et al. Endosomal escape:A bottleneck for LNP-mediated therapeutics[J]. Proceedings of the National Academy of Sciences，2024，121(11):e2307800120.
[19] ZHENG X，ZHENG S，KONG Y，et al. Recent advances in surface defect inspection of industrial products using deep learning techniques[J]. Int J Adv Manuf Technol，2021，113:35-58.
[20] REIS D，KUPEC J，HONG J，et al. Real-time flying object detection with yolov8[J]. arXiv preprint arXiv:2305.09972，2023.
[21] HU J，SHEN L，SUN G，et al. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018:7132-7141.
[22] WANG Q，WU B，ZHU P，et al. Eca-net:Efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020:11531-11539.
[23] WOO S，PARK J，LEE J，et al. CBAM:Convolutional block attention module[C]// Computer Vision - ECCV 2018. Cham:Springer International Publishing，2018:3-19.
[24] PARK J，WOO S，LEE J，et al. BAM:Bottleneck attention module[J]. arXiv preprint arXiv:1807.06514，2018.
[25] HOU Q，ZHOU D Q，FENG J，et al. Coordinate attention for efficient mobile network design[J]. arXiv preprint arXiv:2103.02907，2021.
[26] WANG S，SUN G，ZHENG B，et al. A crop image segmentation and extraction algorithm based on mask RCNN[J]. Entropy，2021，23(9):1160.
[27] CAI Z，VASCONCELOS N. Cascade R-CNN:Delving into high quality object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018:6154-6162.
[28] WANG X，KONG T，SHEN C，et al. Solo:Segmenting objects by locations[C]//Computer Vision-ECCV 2020:16th European Conference，Glasgow，UK，August 23-28，2020，Proceedings，Part XVIII 16. Springer International Publishing，2020:649-665.
[29] IRIE K，NISHIKAWA K. Detection method from 4k images using ssd300 without retraining[C]//2022 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). IEEE，2022:877-883.
[30] TAN M，PANG R，LE Q V. Efficientdet:Scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020:10781-10790.
[31] WANG Y，WANG C，ZHANG H，et al. Automatic ship detection based on RetinaNet using multi-resolution Gaofen-3 imagery[J]. Remote Sensing，2019，11(5):531.
[32] DUAN K，BAI S，XIE L，et al. Centernet:Keypoint triplets for object detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019:6569-6578.
[33] SUN X，WU P，HOI S C H. Face detection using deep learning:An improved faster RCNN approach[J]. Neurocomputing，2018，299:42-50.
[34] BOCHKOVSKIY A，WANG C Y，LIAO H Y M. Yolov4:Optimal speed and accuracy of object detection[J]. arXiv preprint arXiv:2004.10934，2020.
[35] ZHANG M，YE S，ZHAO S，et al. Pear object detection in complex orchard environment based on improved YOLO11[J]. Symmetry，2025，17(2):255.
[36] CHEN S，JIANG S，WANG X，et al. An efficient detector for detecting surface defects on cold-rolled steel strips[J]. Engineering Applications of Artificial Intelligence，2024，138(A):109325.
[37] CHEN S，JIANG S，WANG X，et al. A high-precision and real-time lightweight detection model for small defects in cold-rolled steel[J]. Journal of Real-Time Image Processing，2025，22:30.