Abstract
An improved YOLOv 5 algorithm for bridge apparent disease detection was proposed to address the limitations of target detection algorithms in real bridge modeling disease detection with missing surrounding pixels. The conventional YOLOv 5 model suffers from problems in bridge disease detection, including weak ability in target disease localization and feature extraction, and low recognition accuracy of small-size target diseases. To this end, the CA attention mechanism module was first integrated into the feature extraction backbone network of conventional YOLOv 5 to enhance the model ’s ability to localize and extract features from the target diseases. Secondly, the C 2f module was employed to replace the original CSP module in conventional YOLOv 5 to enhance the model ’s detection accuracy of small-size target diseases. Compared with the conventional YOLOv 5 model, the model embedded with the CA module improves the recognition accuracy of crack and spalling diseases by 2.74% and 2.55%, respectively. By applying the C 2f module on this basis, the model ’s detection accuracy for small-size target diseases of cracks and seepage diseases was further increased by 4.12% and 3.62%, respectively. The best mAP value of the improved YOLOv 5 model on the test set reaches 0.945, which is 1.61% higher than that of conventional YOLOv 5. It shows that the improved YOLOv 5 bridge disease feature extraction and localization ability is stronger, and the detection accuracy of small-size target diseases is higher.
Publication Date
10-29-2026
DOI
10.14048/j.issn.1671-2579.2026.05.027
First Page
257
Last Page
266
Submission Date
October 2026
Recommended Citation
Rong, Huang; Rui, Li; Yuchen, Zhai; and Yan, Lyu
(2026)
"Bridge apparent disease detection based on improved YOLOv 5 algorithm,"
Journal of China & Foreign Highway: Vol. 46:
Iss.
5, Article 27.
DOI: 10.14048/j.issn.1671-2579.2026.05.027
Available at:
https://zwgl1980.csust.edu.cn/journal/vol46/iss5/27
Reference
[1]Wardhana K,Hadipriono F C.Analysis of recent bridge failures in the United States [J].Journal of Performance of Constructed Facilities,2003,17(3):144-150.
[2]贺拴海,赵祥模,马建,等.公路桥梁检测及评价技术综述[J].中国公路学报,2017,30(11):63-80.He Shuanhai,Zhao Xiangmo,Ma Jian,et al.Review of highway bridge inspection and condition assessment [J].China Journal of Highway and Transport,2017,30(11):63-80.
[3]张雨诗,陈国栋,林聪功,等.基于 YOLO-Bridge 的桥梁病害检测方法研究 [J].佳木斯大学学报 (自然科学版 ),2024,42(11):13-17,91.Zhang Yushi,Chen Guodong,Lin Conggong,et al.Research on bridge damage detection method based on YOLO-Bridge [J].Journal of Jiamusi University (Natural Science Edition ),2024,42(11):13-17,91.
[4]宋泽冈,刘艳莉,张长兴.基于机器视觉的桥梁裂缝检测应用及发展综述 [J].科学技术与工程,2023,23(30):12796 -12805.Song Zegang,Liu Yanli,Zhang Changxing.Review on application and development of bridge crack detection based on machine vision [J].Science Technology and Engineering,2023,23(30):12796 -12805.
[5]李佩,韩芳,杨凯,等.基于改进 YOLOv 5s算法的桥梁裂缝自动检测及分类 [J].交通科技,2024 (3):53-58.Li Pei,Han Fang,Yang Kai,et al.Automatic detection and classification of bridge cracks based on improved YOLOv 5s algorithm [J].Transportation Science & Technology,2024 (3):53-58.
[6]黄鹏,范文哲,胡玲玲,等.基于图像连通域的桥梁裂缝识别算法应用研究 [J].中外公路,2025,45(2):126-132.Huang Peng,Fan Wenzhe,Hu Lingling,et al.Application of bridge crack recognition algorithm based on image connected domain [J].Journal of China & Foreign Highway,2025,45(2):126-132.
[7]苏卫国,王景霄.基于 YOLOv 3深度学习算法的道路裂缝识别模型研究 [J].中外公路,2023,43(2):58-63.Su Weiguo,Wang Jingxiao.Research on road crack recognition model based on YOLOv 3 deep learning algorithm [J].Journal of China & Foreign Highway,2023,43(2):58-63.
[8]Jiang S,Cheng Y Y,Zhang J.Vision-guided unmanned aerial system for rapid multiple-type damage detection and localization [J].Structural Health Monitoring,2023,22(1):319-337.[9]Kim B,Natarajan Y,Preethaa K R S,et al.Real-time assessment of surface cracks in concrete structures using integrated deep neural networks with autonomous unmanned aerial vehicle [J].Engineering Applications of Artificial Intelligence,2024,129:107537.
[10]Lyu G Z,Wang P,Li G H,et al.A heavy-load wall-climbing robot for bridge concrete structures inspection[J].Industrial Robot:The International Journal of Robotics Research and Application,2024,51(3):465-478.
[11]Yeum C M,Dyke S J.Vision-based automated crack detection for bridge inspection [J].Computer-Aided Civil and Infrastructure Engineering,2015,30(10):759-770.
[12]Dawood T,Zhu Z H,Zayed T.Machine vision-based model for spalling detection and quantification in subway networks [J].Automation in Construction,2017,81:149-160.
[13]粟寒,斯新华,黄倩文.YOLOv 5算法在混凝土桥梁表观缺陷智能识别的应用研究 [J].公路交通技术,2024,40(6):90-94.Su Han,Si Xinhua,Huang Qianwen.Application research of YOLOv 5 algorithm in intelligent recognition of apparent defects in concrete bridges [J].Technology of Highway and Transport,2024,40(6):90-94.
[14]Chun P J,Izumi S,Yamane T.Automatic detection method of cracks from concrete surface imagery using two-step light gradient boosting machine [J].Computer-Aided Civil and Infrastructure Engineering,2021,36(1):61-72.
[15]Cha Y J,Choi W,Suh G,et al.Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types [J].Computer-Aided Civil and Infrastructure Engineering,2018,33(9):731-747.
[16]Kim I H,Jeon H,Baek S C,et al.Application of crack identification techniques for an aging concrete bridge inspection using an unmanned aerial vehicle [J].Sensors,2018,18(6):1881.
[17]Zhang C B,Chang C C,Jamshidi M.Concrete bridge surface damage detection using a single-stage detector [J].Computer-Aided Civil and Infrastructure Engineering,2020,35(4):389-409.
[18]Xie Y F,Ming S.Bridge defect detection technology based on machine vision and embedded [J].Journal of Physics:Conference Series,2021,1856 (1):012061.
[19]周清松,董绍江,罗家元,等.改进 YOLOv 3的桥梁表观病害检测识别 [J].重庆大学学报,2022,45(6):121-130.Zhou Qingsong,Dong Shaojiang,Luo Jiayuan,et al.Bridge apparent disease detection based on improved YOLOv 3[J].Journal of Chongqing University (Natural Science Edition ),2022,45(6):121-130.
[20]Huang G,Liu Z,Van Der Maaten L,et al.Densely connected convolutional networks [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ).Honolulu,HI,USA.IEEE,2017:2261 -2269.
[21]刘鹏宇,袁静,高倩,等.基于改进 YOLOv 5的路面病害检测方法 [J].北京工业大学 学报,2025,51(5):552-559.Liu Pengyu,Yuan Jing,Gao Qian,et al.Pavement disease detection method based on improved YOLOv 5[J].Journal of Beijing University of Technology,2025,51(5):552-559.
[22]廖延娜,姚亮.改进 YOLOX 的桥梁病害检测识别 [J].应用光学,2023,44(4):792-800.Liao Yanna,Yao Liang.Bridge disease detection and recognition based on improved YOLOX algorithm [J].Journal of Applied Optics,2023,44(4):792-800.
[23]周水兴,罗成,周琳淇.基于 BS-YOLO 11模型的混凝土桥 梁 裂 缝 识 别 研 究 [J].公 路 交 通 技 术,2025,41(2):119-126.Zhou Shuixing,Luo Cheng,Zhou Linqi.Research on crack detection of concrete bridges based on BS-YOLO 11 model [J].Technology of Highway and Transport,2025,41(2):119-126.
[24]Liu Y K,He G L,Wang Z H,et al.NRT-YOLO:Improved YOLOv 5 based on nested residual transformer for tiny remote sensing object detection [J].Sensors,2022,22(13):4953.
[25]王道累,肖贝成,姚从荣,等.基于改进 YOLOv 5的深度学习光伏组件 “热斑效应 ”检测方法 [J].太阳能学报,2024,45(8):342-348.Wang Daolei,Xiao Beicheng,Yao Congrong,et al.Improved YOLOv 5-based deep learning method for detecting “hot spot effect ” in phototvoltaic modules [J].Acta Energiae Solaris Sinica,2024,45(8):342-348.
[26]Hou Q B,Zhou D Q,Feng J S.Coordinate attention for efficient mobile network design [C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR ),2021:13708 -13717.
[27]Mundt M,Majumder S,Murali S,et al.Meta-learning convolutional neural architectures for multi-target concrete defect classification with the concrete defect bridge image dataset [C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR ),2019:11188 -11197.