基于改进Yolov7-tiny的带钢表面缺陷检测算法

doi:10.19678/j.issn.1000-3428.0068397

摘要/Abstract

摘要： 针对带钢表面缺陷检测任务存在小目标检测效率低、缺陷定位不准确、检测算法参数量大、难以运用在终端设备上等问题，提出一种改进的YOLOv7-tiny检测算法。首先，使用GSConv替换颈部网络中的标准卷积，再基于GSConv设计一种改进的高效聚合网络(ELAN-G)，保证带钢表面缺陷特征信息被充分融合的同时降低算法的参数量；其次，在预测头和颈部网络之间增加针对低分辨率和小缺陷的SPDConv模块，模块先生成一个中间特征图，通过对中间特征图中的小缺陷特征信息过滤学习得到最终特征图，以此提高预测头对小缺陷的检测精度；最后，引入MPDIoU损失函数，合理利用边界回归框的几何性质，简化损失函数计算过程并提高缺陷定位精度。实验结果表明，在NEU-DET数据集上，改进算法比其他6种先进目标检测算法效果更好，性能更均衡，改进算法的平均精度均值(mAP)可达74.1％，且参数量和计算量低于所有对比算法，可布置在工业坏境中的带钢表面缺陷检测系统上。

Abstract: To address the problems of low efficiency of small target detection, inaccurate defect localization, large number of parameters in the detection algorithm, and difficult to use the model on the terminal equipment, in the task of detecting surface defects on strip steel, an improved YOLOv7-tiny detection algorithm is proposed. Firstly, the GSConv is used to replace the standard convolution in the Neck network, and then an improved and efficient aggregation network (ELAN-G) is designed based on GSConv, which reduces the parameter amounts of the model while ensuring that the information of the strip steel surface defect features is adequately fused; secondly, the SPDConv module for low-resolution and small defects is added between the Head and the Neck network, and the module is first generated into an intermediate feature map, the final feature map is obtained by filtering and learning the small defect feature information in the intermediate feature map, to improve the detection accuracy of the Head for small defects; finally, the MPDIoU loss function is adopted, and the geometric properties of the bounding regression box are reasonably utilized to simplify the calculation process of the loss function and improve the accuracy of defect localization. The experimental results show that the improved algorithm is better than the other six advanced target detection algorithms on the NEU-DET dataset, with more balanced performance, the mean average accuracy (mAP) of the improved algorithm can reach up to 74.1%, and the parameter amounts and computation is lower than that of all the comparative algorithms, which can be arranged on steel surface defects detection system in the industrial environment.

阳丽莎, 李茂军, 胡建文, 王鼎湘. 基于改进Yolov7-tiny的带钢表面缺陷检测算法[J]. 计算机工程, doi: 10.19678/j.issn.1000-3428.0068397.

Yang Lisha, Li Maojun, Hu Jianwen, Wang Dingxiang. Strip Steel Surface Defect Detection Algorithm based on Improved Yolov7-tiny[J]. Computer Engineering, doi: 10.19678/j.issn.1000-3428.0068397.

参考文献

[1] 李少波, 杨静, 王铮, 朱书德, 杨观赐. 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, Zhu S D, Yang G C. Review of Development and Application of Defect Detection Technology[J]. Acta Automatica Sinica, 2020, 46(11): 2319−2336.
[2] 周亮, 王振环, 孙东辰, 穆乃锋. 现代精密测量技术现状及发展[J]. 仪器仪表学报, 2017, 38(8): 1869−1878. Zhou L, Wang Z H, Sun D C, Mu N F. Present situation and development of modern precision measurement technology[J]. Chinese Journal of Scientific Instrument, 2017, 38(8): 1869−1878.
[3] Park J K, Kwon B K, Park J H, et al. Machine learning based imaging system for surface defect inspection[J]. International Journal of Precision Engineering and Manufacturing-Green Technology, 2016, 3(3): 303-310.
[4] Chu M, Gong R, Gao S, et al. Steel surface defects recognition based on multi-type statistical features and enhanced twin support vector machine[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 171: 140-150.
[5] 赵鹤, 杨晓洪, 李小彤等. 基于贝叶斯网络的铜带表面缺陷图像分类[J]. 控制工程, 2022, 29(10): 2020-1170. Zhao H, Yang X H, Li X T. Image Classification of Copper Strip Surface Defects Based on Bayesian Network[J]. Control Engineering of China, 2022, 29(10): 2020-1170.
[6] Ren S, He K, Girshick R, et al. Faster r-cnn: Towards real-time object detection with region proposal networks[J]. Advances in neural information processing systems, 2015, 28:91-99.
[7] 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[C]//Proceedings of theIEEE/CVF Conference on Computer Vision and Pattern
Recognition. 2023: 7464-7475. [8] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. Springer International Publishing, 2016: 21-37.
[9] Shi X, Zhou S, Tai Y, et al. An Improved Faster R-CNN for Steel Surface Defect Detection[C]//2022 IEEE 24th International Workshop on Multimedia Signal Processing (MMSP). IEEE, 2022: 1-5.
[10] Liu X, Gao J. Surface defect detection method of hot rolling strip based on improved SSD model[C]//Database Systems for Advanced Applications. DASFAA 2021 International Workshops: BDQM, GDMA, MLDLDSA, MobiSocial, and MUST, Taipei, Taiwan, April 11–14, 2021, Proceedings 26. Springer International Publishing, 2021: 209-222.
[11] Zhong H, Wu B, Zhang X, et al. Steel Surface Defect Detection Based on an Improved YOLOv5 Model[C]//2023 5th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP). IEEE, 2023: 51-55.
[12] 曹义亲, 周一纬, 徐露. 基于 E-YOLOX 的实时金属表面缺陷检测算法[J]. 图学学报, 2023, 44(04): 677-690. Cao Y Q, Zhou Y W, Xu L. A real-time metallic surface defect detection algorithm based on E-YOLOX[J]. Journal of Graphics, 2023, 44(04): 677-690.
[13] 窦智, 胡晨光, 李庆华, 郑李明. 改进 YOLOv7 的小样本钢板表面缺陷检测算法[J]. 计算机工程与应用: 1-12. Dou Z, Hu C G, Li Q H, Zheng L M. Improved YOLOv7 Algorithm for Small Sample Steel Plate Surface Defect Detection[J]. Computer Engineering and Applications: 1-12.
[14] He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9): 1904-1916.
[15] Lin T Y, Dollár P, Girshick R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 2117-2125.
[16] Liu S, Qi L, Qin H, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 8759-8768.
[17] Wang C Y, Yeh I H, Liao H Y M. You only learn one representation: Unified network for multiple tasks[J]. arXiv preprint arXiv:2105.04206, 2021.
[18] Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861, 2017.
[19] Zhang X, Zhou X, Lin M, et al. Shufflenet: An extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 6848-6856.
[20] Han K, Wang Y, Tian Q, et al. Ghostnet: More features from cheap operations[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 1580-1589.
[21] Li H, Li J, Wei H, et al. Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles[J]. arXiv preprint arXiv:2206.02424, 2022.
[22] Sunkara R, Luo T. No more strided convolutions or pooling: A new CNN building block for low-resolution images and small objects[C]//Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Cham: Springer Nature Switzerland, 2022: 443-459.
[23] Carion N, Massa F, Synnaeve G, et al. End-to-end object detection with transformers[C]//European conference on computer vision. Cham: Springer International Publishing, 2020: 213-229.
[24] Zheng Z, Wang P, Liu W, et al. Distance-IoU loss: Faster and better learning for bounding box regression[C]//Proceedings of the AAAI conference on artificial intelligence. 2020, 34(07): 12993-13000.
[25] Siliang M, Yong X. MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression[J]. arXiv preprint arXiv:2307.07662, 2023.

选择文件类型/文献管理软件名称

选择包含的内容