YOLOv8-VSC：一种轻量级的带钢表面缺陷检测算法

doi:10.3778/j.issn.1673-9418.2308060

摘要/Abstract

摘要： 目前在带钢表面缺陷检测领域，通用的目标检测算法复杂度高、计算量庞大，而一些中小型企业负责检测的终端设备通常不具备较强的计算能力，计算资源有限，从而导致检测算法部署困难。为解决该问题，基于YOLOv8n目标检测框架，提出一种轻量级的带钢表面缺陷检测模型YOLOv8-VSC。该模型使用轻量级的VanillaNet网络作为骨干特征提取网络，通过减少不必要的分支结构降低模型的复杂度。同时，引入SPD模块在减少网络层数的同时加快模型的推理速度。为了进一步提升检测精度，在特征融合网络中，使用轻量级的上采样算子CARAFE，提高融合特征的质量和丰富度。最后，在NEU-DET数据集上进行大量实验，得到模型的参数量与计算量为1.96×106和6.0 GFLOPs，仅为基线的65.1%和74.1%，mAP达到80.8%，较基线提升1.8个百分点。此外，在铝材表面缺陷数据集和VOC2012数据集上的实验结果表明所提算法具有良好的鲁棒性。与先进的目标检测算法相比，所提算法在保证高检测精度的前提下需要的计算资源更少。

关键词: 缺陷检测, 带钢表面缺陷, YOLOv8, 轻量级网络, VanillaNet

Abstract: Currently, in the field of strip steel surface defect detection, the generalized target detection algorithm is highly complex and computationally large, while terminal equipment responsible for the detection of some small and medium-sized enterprises usually does not have strong computational capabilities, and the computational resources are limited, which leads to difficulties in the deployment of detection algorithms. To solve this problem, this paper proposes a lightweight strip steel surface defect detection model YOLOv8-VSC based on the YOLOv8n target detec-tion framework, which uses a lightweight VanillaNet network as the backbone feature extraction network and reduces the complexity of the model by reducing the unnecessary branching structure. Meanwhile, the SPD module is introduced to speed up the inference of the model while reducing the number of network layers. To further improve the detection accuracy, a lightweight up-sampling operator, CARAFE, is used in the feature fusion network to improve the quality and richness of the features. Finally, extensive experiments on the NEU-DET dataset yield a model with parametric and computational quantities of 1.96×106 and 6.0 GFLOPs, which are only 65.1% and 74.1% of the baseline, and the mAP reaches 80.8%, which is an improvement of 1.8 percentage points from the baseline. In addition, experimental results on the aluminum surface defect dataset and the VOC2012 dataset show that the proposed algorithm has good robustness. Compared with advanced target detection algorithms, the proposed algorithm requires fewer computational resources while ensuring high detection accuracy.

Key words: defect detection, strip surface defect, YOLOv8, lightweight network, VanillaNet

王春梅, 刘欢. YOLOv8-VSC：一种轻量级的带钢表面缺陷检测算法[J]. 计算机科学与探索, 2024, 18(1): 151-160.

WANG Chunmei, LIU Huan. YOLOv8-VSC: Lightweight Algorithm for Strip Surface Defect Detection[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(1): 151-160.

参考文献

[1] JIAN C, GAO J, AO Y. Automatic surface defect detection for mobile phone screen glass based on machine vision[J]. Applied Soft Computing, 2017, 52: 348-358.
[2] HE Z, LIU Q. Deep regression neural network for industrial surface defect detection[J]. IEEE Access, 2020, 8: 35583-35591.
[3] LIU Y, XIAO H, XU J, et al. A rail surface defect detection method based on pyramid feature and lightweight convo-lutional neural network[J]. IEEE Transactions on Instrum-entation and Measurement, 2022, 71: 1-10.
[4] SINGH S A, DESAI K A. Automated surface defect detection framework using machine vision and convolutional neural networks[J]. Journal of Intelligent Manufacturing, 2023, 34(4): 1995-2011.
[5] LIANG F T, ZHOU Y, CHEN X, et al. Review of target detection technology based on deep learning[C]//Proce-edings of the 5th International Conference on Control Eng-ineering and Artificial Intelligence. New York: ACM, 2021: 132-135.
[6] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Washington: IEEE Computer Society, 2015: 1440-1448.
[7] ZHANG K H, SHEN H K. Solder joint defect detection in the connectors using improved faster-RCNN algorithm[J]. Applied Sciences, 2021, 11(2): 576.
[8] YANG A M, JIANG T Y, HAN Y, et al. Research on appli-cation of on-line melting in-situ visual inspection of iron ore powder based on Faster R-CNN[J]. Alexandria Engine-ering Journal, 2022, 61(11): 8963-8971.
[9] KUMAR A, MANIKANDAN R. Brain tumor detection using deep neural network-based classifier[C]//Proceedings of the 2022 International Conference on Innovative Computing and Communications. Singapore: Springer, 2022: 173-181.
[10] GUO F, QIAN Y, RIZOS D, et al. Automatic rail surface defects inspection based on Mask R-CNN[J]. Transport-ation Research Record, 2021(11): 655-668.
[11] REN J S, WANG Y. Overview of object detection algor-ithms using convolutional neural networks[J]. Journal of Computer and Communications, 2022, 10(1): 115-132.
[12] JIANG P Y, ERGU D, LIU F Y, et al. A review of YOLO algorithm developments[J]. Procedia Computer Science, 2022, 199: 1066-1073.
[13] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//LNCS 12346: Proce-edings of the 16th European Conference on Computer Vision. Cham: Springer, 2020: 213-229.
[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.
CAI J F, BAI J J, ZHANG X, et al. Research on surface defect recognition of metal sheet based on improved Mask R-CNN[J]. Journal of Chongqing Institute of Science and Technology (Natural Science Edition), 2023, 25(2): 110-116.
[15] ZHOU S, ZENG Y, LI S, et al. Surface defect detection of rolled steel based on lightweight model[J]. Applied Sciences, 2022, 12(17): 8905.
[16] YANG L, HUANG X, REN Y, et al. Steel plate surface defect detection based on dataset enhancement and light-weight convolution neural network[J]. Machines, 2022, 10(7): 523.
[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.
ZHANG Z C. Lightweight strip steel defect detection based on improved YOLOv5[J]. Computer System Applications, 2023, 32(6): 278-285.
[18] QIN R, CHEN N, HUANG Y. EDDNet: an efficient and accurate defect detection network for the industrial edge environment[C]//Proceedings of the 2022 IEEE 22nd Inter-national Conference on Software Quality, Reliability and Security. Piscataway: IEEE, 2022: 854-863.
[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.
YAN X, YANG Y C, TU N W. Steel surface defect detec-tion based on improved SSD[J]. Modern Manufacturing Engineering, 2023(5): 112-120.
[20] 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[J]. 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 envi-ronments[J]. Computer Engineering and Applications, 2023, 59(15): 318-328.
[21] WANG J, XU P, LI L, et al. DAssd-Net: a lightweight steel surface defect detection model based on multi-branch dilated convolution aggregation and multi-domain perception detec-tion head[J]. Sensors, 2023, 23(12): 5488.
[22] 周颖, 颜毓泽, 陈海永, 等. 基于改进YOLOv8的光伏电池缺陷检测[J]. 激光与光电子学进展, 2024, 61(8): 8.
ZHOU Y, YAN Y Z, CHEN H Y, et al. Defect detection of photovoltaic cells based on improved YOLOv8[J]. Adva-nces in Lasers and Optoelectronics, 2024, 61(8): 8.
[23] PHAN Q B, NGUYEN T. A novel approach for PV cell fault detection using YOLOv8 and particle swarm optimiz-ation[R/OL]. [2023-07-16]. https://doi.org/10.36227/techrxiv. 22680484.v1.
[24] CHEN H, WANG Y, GUO J, et al. VanillaNet: the power of minimalism in deep learning[J]. arXiv:2305.12972, 2023.
[25] SUNKARA R, LUO T. No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects[C]//Proceedings of the 2022 Joint Euro-pean Conference on Machine Learning and Knowledge Discovery in Databases. Cham: Springer, 2022: 443-459.
[26] WANG J, CHEN K, XU R, et al. CARAFE: content-aware reassembly of features[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscat-away: IEEE, 2019: 3007-3016.
[27] LI X, WANG C, JU H, et al. Surface defect detection model for aero-engine components based on improved YOLOv5[J]. Applied Sciences, 2022, 12(14): 7235.
[28] CHEN M, YU L, ZHI C, et al. Improved faster R-CNN for fabric defect detection based on Gabor filter with genetic algorithm optimization[J]. Computers in Industry, 2022, 134: 103551.
[29] DANG L M, WANG H, LI Y, et al. DefectTR: end-to-end defect detection for sewage networks using a transformer[J]. Construction and Building Materials, 2022, 325: 126584.
[30] CUI L, JIANG X, XU M, et al. SDDNet: a fast and accurate network for surface defect detection[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-13.
[31] ZHANG L, YAN S, HONG J, et al. An improved defect recognition framework for casting based on DETR algor-ithm[J]. Journal of Iron and Steel Research International, 2023, 30(5): 949-959.
[32] SURYARASMI A, CHANG C C, AKHMALIA R, et al. FN-Net: a lightweight CNN-based architecture for fabric defect detection with adaptive threshold-based class determination[J]. Displays, 2022, 73: 102241.
[33] DING K, NIU Z, HUI J, et al. A weld surface defect reco-gnition method based on improved MobileNetV2 algorithm[J]. Mathematics, 2022, 10(19): 3678.
[34] SHI C, LIN L, SUN J, et al. A lightweight YOLOv5 trans-mission line defect detection method based on coordinate attention[C]//Proceedings of the 2022 IEEE 6th Inform-ation Technology and Mechatronics Engineering Conference. Piscataway: IEEE, 2022: 1779-1785.
[35] MA Z, LI Y, HUANG M, et al. Automated real-time dete-ction of surface defects in manufacturing processes of aluminum alloy strip using a lightweight network architec-ture[J]. Journal of Intelligent Manufacturing, 2023, 34(5): 2431-2447.
[36] WANG H, YANG X, ZHOU B, et al. Strip surface defect detection algorithm based on YOLOv5[J]. Materials, 2023, 16(7): 2811.
[37] ZHOU X, WANG D, KR?HENBüHL P. Objects as points[J]. arXiv:1904.07850, 2019.