融合Res2Net和部分卷积的带钢表面缺陷检测算法

doi:10.3778/j.issn.1002-8331.2309-0234

摘要/Abstract

摘要： 为提升带钢表面缺陷的层次性特征提取能力和检测效率，提出基于多尺度表征和部分卷积（PConv）的快速检测网络（multi-scale and partially convolutional network，MSPC-Net）。将限制对比度自适应直方图均衡技术（contrast limited adaptive histogram equalization，CLAHE）引入到该模型以突出带钢表面的缺陷特征；在YOLOv5s的基础上新增检测层，提高对不同尺寸缺陷目标的检测率；设计了融合Res2Net的多尺度特征提取块并引入ECA注意力机制（BRE-block），既可以获取细粒度层面的特征，同时也增加了模型感受野；通过结合PConv减少了模型计算量（FLOPs），且增强了部分特征信息的聚合。在NEU-DET数据集上的实验结果表明，平均精度均值（mAP@IoU=0.5）达到了80.2%，较原基线网络提高了5.9个百分点；同时改进后网络的FPS达到157，远高于近期应用广泛的目标检测算法，有效提高了带钢表面缺陷的检测效率。

关键词: 对比度增强, 多尺度特征提取, 改进Res2Net, 融合PConv, 信息聚合

Abstract: In order to improve the hierarchical feature extraction capability and detection efficiency of strip surface defects, a rapid detection network (MSPC-Net) based on multi-scale representation and partial convolution (PConv) is proposed. Contrast limited adaptive histogram equalization (CLAHE) technology is introduced into the model to highlight the defect characteristics of the strip surface; a new detection layer is added based on YOLOv5s to improve the detection rate of defect targets of different sizes. A multi-scale feature extraction block that integrates Res2Net is designed, and the ECA attention mechanism (BRE-block) is introduced. This block can not only obtain fine-grained features but also increase the model receptive field. By combining with PConv (FLOPs), the amount of model calculation is reduced, and the aggregation of partial feature information is enhanced. Experimental results on the NEU-DET data set show that the average accuracy (mAP@IoU=0.5) reaches 80.2%, which is 5.9 percentage points higher than the original baseline network. At the same time, the FPS of the improved network reaches 157, which is much higher than that of the recently widely used target detection algorithm, effectively improving the detection efficiency of strip surface defects.

Key words: contrast enhancement, multi-scale feature extraction, improved Res2Net, fusion of PConv, information aggregation

胡凯涛, 马向华, 孙向宇, 刘闯. 融合Res2Net和部分卷积的带钢表面缺陷检测算法[J]. 计算机工程与应用, 2025, 61(5): 334-343.

HU Kaitao, MA Xianghua, SUN Xiangyu, LIU Chuang. Strip Surface Defect Detection Algorithm Integrating Res2Net and PConv[J]. Computer Engineering and Applications, 2025, 61(5): 334-343.

参考文献

[1] 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.
[2] WANG Y, LIAN X, YE L. Product defects detection based on features fusion and the multi-scale sliding window[C]//Proceedings of the 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, 2017: 2948-2952.
[3] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
[4] GIRSHICK R. Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440-1448.
[5] REN S, HE K, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems, 2017: 1137-1149.
[6] REDMON J, DIVVALA S, GIRSHICK R, et al. You only lookonce: unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779-788.
[7] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of the Computer 14th European Conference on Vision, 2016: 21-37.
[8] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. Yolov4: optimal speed and accuracy of object detection[J]. arXiv: 2004.10934, 2020.
[9] WANG C Y, LIAO H Y M, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020: 390-391.
[10] 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.
CHEMG 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.
[11] QI J, LIU X, LIU K, et al. An improved YOLOv5 model based on visual attention mechanism: application to recognition of tomato virus disease[J]. Computers and Electronics in Agriculture, 2022, 194: 106780.
[12] LI Z, TIAN X, LIU X, et al. A two-stage industrial defect detection framework based on improved-yolov5 and optimized-inception-resnetv2 models[J]. Applied Sciences, 2022, 12(2): 834.
[13] CHEN H, DU Y, FU Y, et al. DCAM-Net: a rapid detection network for strip steel surface defects based on deformable convolution and attention mechanism[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-12.
[14] ZHANG J, KANG X, NI H, et al. Surface defect detection of steel strips based on classification priority YOLOv3-dense network[J]. Ironmaking & Steelmaking, 2021, 48(5): 547-558.
[15] WANG H, YANG X, ZHOU B, et al. Strip surface defect detection algorithm based on YOLOv5[J]. Materials, 2023, 16(7): 2811
[16] 彭晏飞, 袁晓龙, 陈炎康, 等. 改进YOLOv5s的带钢表面缺陷检测方法[J/OL]. 机械科学与技术: 1-9[2023-09-22]. https: //doi.org/10.13433/j.cnki.1003-8728.20230254.
PENG Y F, YUAN X L, CHEN Y K, et al. Strip steel surface defect detection method by improved YOLOv5 network[J/OL]. Mechanical Science and Technology: 1-9 [2023-09-22]. https://doi.org/10.13433/j.cnki.1003-8728.20230254.
[17] YANG N, GUO W. Application of improved yolov5 model for strip surface defect detection[C]//Proceedings of the 2022 Global Reliability and Prognostics and Health Management, 2022: 1-5.
[18] LI J, SU Z, GENG J, et al. Real-time detection of steel strip surface defects based on improved YOLO detection network[J]. IFAC-Papers OnLine, 2018, 51(21): 76-81.
[19] NATH V, CHATTOPADHYAY C, DESAI K A. On enhancing prediction abilities of vision-based metallic surface defect classification through adversarial training[J]. Engineering Applications of Artificial Intelligence, 2023, 117: 105553.
[20] GAO S H, CHENG M M, ZHAO K, et al. Res2Net: a new multi-scale backbone architecture[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 43(2): 652-662.
[21] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[22] WANG Q, WU B, ZHU P, et al. ECA-Net: efficient channel attentionfor deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11534-11542.
[23] CHEN J, KAO S, HE H, et al. Run, don’t walk: chasing HigherFLOPS for faster neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 12021-12031.
[24] SONG K, YAN Y. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects[J]. Applied Surface Science, 2013, 285: 858-864.