结合局部强化和改进YOLOv8的隧道螺栓锈蚀检测

doi:10.3778/j.issn.1002-8331.2406-0288

摘要/Abstract

摘要： 针对地铁隧道检修环境光线不足，不可避免地导致人工检修螺栓准确率低、漏诊率高的问题，提出了基于局部强化（local enhancement algorithm，LEA）和改进YOLOv8的隧道螺栓锈蚀检测模型（YOLOv8s+LEA+MSSf+FL，YOLO-LMF），将人工检修变为智能检测，提高检修效率。使用带有邻域检查（neighbor check）的局部强化算法增强螺栓锈蚀部位，使模型更好识别锈蚀特征。提出多尺度通道组混排卷积（multi-scale channel group shuffle convolution，MSCGSC），将MSCGSC融入YOLOv8的C2f（cross stage partial network fusion）模块中，得到新的模块MSSf（multi-scale shuffle fusion），使模型更好地学习锈蚀螺栓与色斑在螺栓附近时的不同的表现，提高模型检测精度。考虑到锈蚀螺栓中困难样本限制了模型检测的精度且螺栓样本不平衡的问题，引入了焦点损失函数（focal loss，FL），降低数量庞大的样本在训练中所占的权重，使模型集中对分类困难样本的学习。实验结果表明：所提出的模型相较于原模型分别增长了0.032、0.05、0.011和0.003，参数量减少了10.4%。模型在地铁隧道螺栓数据集上具有更好的表现，能够为地铁隧道维护作业研发检测机器人提供参考，减少隧道养护工人工作量，提高工作效率。

关键词: 隧道螺栓, 局部强化, YOLOv8, 多尺度通道组混排卷积, 焦点损失

Abstract: A tunnel bolt corrosion detection model based on local enhancement algorithm (LEA) and improved YOLOv8(YOLOv8s+LEA+MSSf+FL, YOLO-LMF) is proposed to address the problem of low accuracy and high missed diagnosis rate of manual bolt inspection due to insufficient lighting in the maintenance environment of subway tunnels, transforming manual maintenance into intelligent detection to improve maintenance efficiency. Firstly, a local enhancement algorithm with neighbor check is used to enhance the bolt corrosion location, enabling the model to better identify corrosion features. Secondly, multi-scale channel group shuffle convolution (MSCGSC) is proposed. Integrating MSCGSC into the C2f (cross stage partial network fusion) module of YOLOv8, a new module MSSf (multi-scale shuffle fusion) is obtained, which enables the model to better learn the different behaviors of corroded bolts and stains near the bolts, and improve the detection accuracy of the model. Finally, considering the limitation of difficult samples in corroded bolts on the accuracy of model detection and the problem of imbalanced bolt samples, focal loss (FL) is introduced to reduce the weight of a large number of samples in training, allowing the model to concentrate on learning difficult samples for classification. The results show that the proposed model has increased by 0.032, 0.05, 0.011, and 0.003 respectively compared to the original model, and the number of parameters has decreased by 10.4%. The model performs better on the bolt dataset of subway tunnels, providing reference for the development of inspection robots for subway tunnel maintenance operations, reducing the workload of tunnel maintenance workers, and improving work efficiency.

Key words: tunnel bolts, local enhancement, YOLOv8, multi-scale channel group shuffle convolution (MSCGSC), focal loss

武晓春, 李鲁豫. 结合局部强化和改进YOLOv8的隧道螺栓锈蚀检测[J]. 计算机工程与应用, 2025, 61(19): 249-259.

WU Xiaochun, LI Luyu. Tunnel Bolt Rust Detection Combining Local Enhancement and Improvement of YOLOv8[J]. Computer Engineering and Applications, 2025, 61(19): 249-259.

参考文献

[1] 张雨洁, 闫海峰, 骆泳吉, 等. 考虑动态随机客流的城市轨道交通列车时刻表仿真优化[J]. 铁道运输与经济, 2024, 46(5): 161-170.
ZHANG Y J, YAN H F, LUO Y J, et al. Simulation optimization of urban rail transit train timetable based on dynamic random passenger flow[J]. Railway Transport and Economy, 2024, 46(5): 161-170.
[2] 周华飞, 梁一鸣. 基于虚拟传感器的结构位移视觉测量误差抑制方法[J]. 浙江工业大学学报, 2024, 52(3): 305-311.
ZHOU H F, LIANG Y M. Virtual sensor-based error suppression method for visual measurement of structural displacement[J]. Journal of Zhejiang University of Technology, 2024, 52(3): 305-311.
[3] 彭大鑫, 甄彤, 李智慧. 低光照图像增强研究方法综述[J]. 计算机工程与应用, 2023, 59(18): 14-27.
PENG D X, ZHEN T, LI Z H. Survey of research methods for low light image enhancement[J]. Computer Engineering and Applications, 2023, 59(18): 14-27.
[4] 卫依雪, 周冬明, 王长城, 等. 结合多分支结构和U-net的低照度图像增强[J]. 计算机工程与应用, 2022, 58(12): 199-208.
WEI Y X, ZHOU D M, WANG C C, et al. Low-light image enhancement using multi-branch structure and U-net[J]. Computer Engineering and Applications, 2022, 58(12): 199-208.
[5] 隋涛, 吴森炜, 贾浩, 等. 基于改进循环生成对抗网络的低照度图像增强[J]. 科学技术与工程, 2024, 24(14): 5911-5919.
SUI T, WU S W, JIA H, et al. Low-light image enhancement based on improved cycle generative adversarial network[J]. Science Technology and Engineering, 2024, 24(14): 5911-5919.
[6] TAN L, TANG T, YUAN D J. An ensemble learning aided computer vision method with advanced color enhancement for corroded bolt detection in tunnels[J]. Sensors, 2022, 22(24): 9715.
[7] 赵晓宇, 陈建军, 张凯琪, 等. 基于HSV色彩空间和Otsu算法的无人机影像植被覆盖度自动提取[J]. 科学技术与工程, 2021, 21(35): 15160-15166.
ZHAO X Y, CHEN J J, ZHANG K Q, et al. Automatic extraction of vegetation coverage from unmanned aerial vehicle images based on HSV and otsu algorithm[J]. Science Technology and Engineering, 2021, 21(35): 15160-15166.
[8] 顾文娟, 丁灿, 魏金, 等. 基于双边滤波MSR与AutoMSRCR融合的低光照图像增强[J]. 光学精密工程, 2023, 31(24): 3606-3617.
GU W J, DING C, WEI J, et al. Low-light image enhancement based on the fusion of Bilateral filter MSR and Auto-
MSRCR[J]. Optics and Precision Engineering, 2023, 31(24): 3606-3617.
[9] 曹超, 顾幸生. 基于改进YOLO v8的行李追踪技术[J]. 计算机工程与应用, 2024, 60(9): 151-158.
CAO C, GU X S. Baggage tracking technology based on improved YOLO v8[J]. Computer Engineering and Applications, 2024, 60(9): 151-158.
[10] 王磊, 郝涌汀, 潘明然, 等. 电力巡检中改进YOLOv5s的缺陷检测算法研究[J]. 计算机工程与应用, 2024, 60(10): 256-265.
WANG L, HAO Y T, PAN M R, et al. Improved defect detection algorithm in power inspection based on YOLOv5s[J]. Computer Engineering and Applications, 2024, 60(10): 256-265.
[11] 张利丰, 田莹. 改进YOLOv8的多尺度轻量型车辆目标检测算法[J]. 计算机工程与应用, 2024, 60(3): 129-137.
ZHANG L F, TIAN Y. Improved YOLOv8 multi-scale and lightweight vehicle object detection algorithm[J]. Computer Engineering and Applications, 2024, 60(3): 129-137.
[12] TAN L, CHEN X H, YUAN D J, et al. DSNet: a computer vision-based detection and corrosion segmentation network for corroded bolt detection in tunnel[J]. Structural Control and Health Monitoring, 2024, 2024(1): 1898088.
[13] OH Y, JEON M, KO D, et al. Randomly shuffled convolution for self-supervised representation learning[J]. Information Sciences, 2023, 623: 206-219.
[14] 冯勇, 杨思卓, 徐红艳. 基于YOLO v8的轻量化安全帽佩戴检测算法[J]. 计算机应用, 2024, 44(S2): 251-256.
FENG Y, YANG S Z, XU H Y. Detection algorithm of lightweight helmet wearing based on YOLO v8[J]. Journal of Computer Applications, 2024, 44(S2): 251-256.
[15] 刘传洋, 吴一全, 刘景景. 基于视觉的输电线路金具锈蚀缺陷检测方法研究进展[J]. 仪器仪表学报, 2024, 45(3): 286-305.
LIU C Y, WU Y Q, LIU J J. Research progress of vision-based rust defect detection methods for metal fittings in transmission lines[J]. Chinese Journal of Scientific Instrument, 2024, 45(3): 286-305.
[16] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 770-778.
[17] ZHANG X Y, ZHOU X Y, LIN M X, et al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 6848-6856.
[18] SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2015: 1-9.
[19] 谢焕增, 黄凌泽, 罗烨, 等. 基于多尺度卷积神经网络的CRISPR/Cas9脱靶预测方法[J]. 生物工程学报, 2024, 40(3): 858-876.
XIE H Z, HUANG L Z, LUO Y, et al. Prediction of CRISPR/Cas9 off-target activity using multi-scale convolutional neural network[J]. Chinese Journal of Biotechnology, 2024, 40(3): 858-876.
[20] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 2999-3007.
[21] FUNT B V, CIUREA F, MCCANN J J. Retinex in MATLABTM[J]. Journal of Electronic Imaging, 2004, 13(1): 48-57.
[22] 高民, 陈高华, 古佳欣, 等. FLM-YOLOv8: 一种轻量级的口罩佩戴检测算法[J]. 计算机工程与应用, 2024, 60(17): 203-215.
GAO M, CHEN G H, GU J X, et al. FLM-YOLOv8: lightweight mask wearing detection algorithm[J]. Computer Engineering and Applications, 2024, 60(17): 203-215.
[23] ZHENG Q H, TIAN X Y, YU Z G, et al. MobileRaT: a lightweight radio transformer method for automatic modulation classification in drone communication systems[J]. Drones, 2023, 7(10): 596.
[24] ZHENG Q H, SAPONARA S, TIAN X Y, et al. A real-time constellation image classification method of wireless communication signals based on the lightweight network MobileViT[J]. Cognitive Neurodynamics, 2024, 18(2): 659-671.
[25] TAN L, HU X X, TANG T, et al. A lightweight metro tunnel water leakage identification algorithm via machine vision[J]. Engineering Failure Analysis, 2023, 150: 107327.
[26] HU X X, CAO Y, SUN Y K, et al. Railway automatic switch stationary contacts wear detection under few-shot occasions[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 14893-14907.
[27] TAN L, CHEN X H, HU X X, et al. DMDSNet: a computer vision-based dual multi-task model for tunnel bolt detection and corrosion segmentation[C]//Proceedings of the 2023 IEEE 26th International Conference on Intelligent Transportation Systems. Piscataway: IEEE, 2023: 4827-4833.
[28] 李善军, 梁千月, 余勇华, 等. 柑橘木虱YOLO v8-MC识别算法与虫情远程监测系统研究[J]. 农业机械学报, 2024, 55(6): 210-218.
LI S J, LIANG Q Y, YU Y H, et al. Research on Asian citrus psyllid YOLO v8-MC recognition algorithm and insect remote monitoring system[J]. Transactions of the Chinese Society for Agricultural Machinery, 2024, 55(6): 210-218.