基于计算机视觉的钢结构螺栓病害智能检测方法研究
Research on Intelligent Detection Methods for Bolt Defects in Steel Structures Based on Computer Vision
针对钢结构螺栓病害检测中人工目检效率低、漏检和误判率高,传统深度学习模型检测精度差、抗噪能力弱且计算参数量大等问题,提出了一种基于改进YOLOv8的螺栓病害检测模型。该模型采用多尺度卷积改进骨干网络,通过多膨胀率卷积核增强了空间特征提取能力;通过引入CSPStage和DySample模块优化颈部网络,强化了多尺度特征融合的效率和质量,提高了模型在复杂场景下的抗噪能力。数据集获取方面利用DJI Mavic 4 Pro无人机采集了平阴黄河大桥钢桁架节点螺栓图像,结合CLAHE预处理与数据增强策略提升图像数据质量,构建了包含锈蚀、松动、缺失三类螺栓病害和正常状态的钢结构螺栓检测数据集。实验结果表明,相同参数设置下,改进后的YOLOv8-GDFPN模型在自建数据集上平均准确率均值和准确率分别为83.5%和88.1%,较YOLOv8模型分别提升了4.9%和4.5%,模型计算参数量为3.263 M,FPS值为181.8 帧·s-1,实现了检测精度与推理效率的平衡,该模型可为户外复杂场景下的螺栓病害智能检测提供参考。
To address issues in steel structure bolt defect detection such as low efficiency, high missed detection and false judgment rate of manual visual inspection, poor detection accuracy, weak anti-noise capability, and large number of computational parameters of traditional deep learning models, this paper proposes a bolt defect detection model based on improved YOLOv8. The model improves the backbone using poly-scale convolution and enhances spatial feature extraction capability through multi-dilation rate convolution kernels. It introduces CSPStage and DySample modules to optimize the neck, strengthens the efficiency and quality of multi-scale feature fusion, and improves the model's anti-noise capability in complex scenarios. For dataset acquisition, DJI Mavic 4 Pro unmanned aerial vehicle was used to collect images of steel truss joint bolts from the Pingyin Yellow River Bridge. Combined with CLAHE preprocessing and data augmentation strategies to improve image data quality, a steel structure bolt defect detection dataset containing four states: rusty, loose, missing, and normal was constructed. Experimental results show that the improved YOLOv8-GDFPN model achieves a mean average precision (mAP) of 83.5% and an accuracy of 88.1% on the self-built dataset, which are 4.9% and 4.5% higher than the YOLOv8 model, respectively. The model has 3.263 M computational parameters and an FPS value of 181.8 frames per second, achieving a balance between detection accuracy and inference efficiency. This model can provide a reference for intelligent detection of bolt defects in complex outdoor scenarios.
| [1] |
孙虎, 肖达, 侯俊. 螺栓连接结构孔边损伤定量化监测技术研究进展[J]. 航空科学技术, 2025, 36(7): 8-18. DOI:10.19452/j.issn1007-5453.2025.07.002. |
| [2] |
SUN Hu, XIAO Da, HOU Jun. Research progress of hole-edge damage quantification technology for bolted joints[J]. Aeronautical Science & Technology, 2025, 36(7): 8-18. DOI:10.19452/j.issn1007-5453.2025.07.002.(in Chinese) |
| [3] |
朱亚飞, 任伟新, 王亚飞. 基于桥梁监测系统振动响应的浮吊船-悬索桥主梁碰撞荷载估算与分析[J]. 中国公路学报, 2024, 37(9): 107-118. DOI:10.19721/j.cnki.1001-7372.2024.09.009. |
| [4] |
ZHU Yafei, REN Weixin, WANG Yafei. Impact force estimation and analysis of collision between floating crane and main girder of suspension bridge using vibration responses in bridge monitoring system[J]. China Journal of Highway and Transport, 2024, 37(9): 107-118. DOI:10.19721/j.cnki.1001-7372.2024.09.009.(in Chinese) |
| [5] |
鞠定超, 张云开, 黄永, 面向智能运维的交通基础设施视觉感知与诊断[J]. 土木工程与管理学报, 2025, 42(3): 1-15. DOI:10.13579/j.cnki.2095-0985.2025.20250087. |
| [6] |
JU Dingchao, ZHANG Yunkai, HUANG Yong, et al. Visual perception and diagnosis of transportation infrastructure for intelligent operation and maintenance[J]. Journal of Civil Engineering and Management, 2025, 42(3): 1-15. DOI:10.13579/j.cnki.2095-0985.2025.20250087.(in Chinese) |
| [7] |
崔闯, 罗纯坤, 邱师津, 基于数据深度增强的钢桥螺栓脱落智能检测方法研究[J]. 桥梁建设, 2024, 54(2): 39-47. DOI:10.20051/j.issn.1003-4722.2024.02.006. |
| [8] |
CUI Chuang, LUO Chunkun, QIU Shijin, et al. An intelligent bolt shedding detection method for steel bridges based on depth data enhancement[J]. Bridge Construction, 2024, 54(2): 39-47. DOI:10.20051/j.issn.1003-4722.2024.02.006.(in Chinese) |
| [9] |
CHA Y J, YOU K, CHOI W. Vision-based detection of loosened bolts using the Hough transform and support vector machines[J]. Automation in Construction, 2016, 71: 181-188. DOI:10.1016/j.autcon.2016.06.008. |
| [10] |
FORKAN A R M, KANG Y B, JAYARAMAN P P, et al. CorrDetector: A framework for structural corrosion detection from drone images using ensemble deep learning[J]. Expert Systems with Applications, 2022, 193: 116461. DOI:10.1016/j.eswa.2021.116461. |
| [11] |
江旺玉, 王乐, 姚叶鹏, 多尺度特征聚合扩散和边缘信息增强的小目标检测算法[J]. 计算机工程与应用, 2025, 61(7): 105-116. |
| [12] |
JIANG Wangyu, WANG Le, YAO Yepeng, et al. Multi-scale feature aggregation diffusion and edge information enhancement small object detection algorithm[J]. Computer Engineering and Applications, 2025, 61(7): 105-116. (in Chinese) |
| [13] |
马耀安, 黄裕婷, 张健豪, 基于K-means-CNN的复杂牌号混合合金样品分类研究[J]. 光谱学与光谱分析, 2025, 45(7): 1946-1952. DOI:10.3964/j.issn.1000-0593(2025)07-1946-07. |
| [14] |
MA Yaoan, HUANG Yuting, ZHANG Jianhao, et al. K-means-CNN-based classification study of mixed alloy samples of complex grades[J]. Spectroscopy and Spectral Analysis, 2025, 45(7): 1946-1952. DOI:10.3964/j.issn.1000-0593(2025)07-1946-07.(in Chinese) |
| [15] |
韦超, 钱春雨, 黄启鹏, 基于YOLOv8n的表格线检测改进模型[J]. 计算机工程与应用, 2025, 61(2): 112-123. DOI:10.3778/j.issn.1002-8331.2405-0165. |
| [16] |
WEI Chao, QIAN Chunyu, HUANG Qipeng, et al. Improved model for table-line detection based on YOLOv8n[J]. Computer Engineering and Applications, 2025, 61(2): 112-123. DOI:10.3778/j.issn.1002-8331.2405-0165.(in Chinese) |
| [17] |
沈浩, 江臣, 陈宇文, 基于深度学习的钢桁架桥螺栓病害智能识别方法[J]. 南京工业大学学报(自然科学版), 2020, 42(5): 608-615. DOI:10.3969/j.issn.1671-7627.2020.05.009. |
| [18] |
SHEN Hao, JIANG Chen, CHEN Yuwen, et al. Intelligent identification method for bolt diseases of steel truss bridge based on deep learning model[J]. Journal of Nanjing University of Technology (Natural Science Edition), 2020, 42(5): 608-615. DOI:10.3969/j.issn.1671-7627.2020.05.009.(in Chinese) |
| [19] |
宫涛, 杨建华, 庄絮竹, 螺栓松动故障监测实验研究综述[J]. 机械设计与制造, 2024(2): 354-363. DOI:10.19356/j.cnki.1001-3997.20230824.019. |
| [20] |
GONG Tao, YANG Jianhua, ZHUANG Xuzhu, et al. Review of experimental research on bolt looseness monitoring[J]. Machinery Design & Manufacture, 2024(2): 354-363. DOI:10.19356/j.cnki.1001-3997.20230824.019.(in Chinese) |
| [21] |
SHANG Z Q, QIN X, ZHANG Z J, et al. Bolt loosening and preload loss detection technology based on machine vision[J]. Buildings, 2024, 14(12): 3897. DOI:10.3390/buildings14123897. |
| [22] |
冒建亮, 王展, 周昕, 基于深度强化学习的机械臂动态避障算法设计与实验验证[J]. 实验技术与管理, 2025, 42(4): 78-85. DOI:10.16791/j.cnki.sjg.2025.04.010. |
| [23] |
MAO Jianliang, WANG Zhan, ZHOU Xin, et al. Design and experimental verification of a dynamic obstacle avoidance algorithm for robot manipulators based on deep reinforcement learning[J]. Experimental Technology and Management, 2025, 42(4): 78-85. DOI:10.16791/j.cnki.sjg.2025.04.010.(in Chinese) |
| [24] |
徐刚年, 王来永, 李伟, 胜利黄河大桥病害高强螺栓节点修复及桥面板补强技术[J]. 桥梁建设, 2023, 53(6): 151-158. DOI:10.20051/j.issn.1003-4722.2023.06.020. |
| [25] |
XU Gangnian, WANG Laiyong, LI Wei, et al. Rehabilitation of damaged high-strength bolted joints and bridge deck strengthening techniques for Shengli Huanghe River bridge[J]. Bridge Construction, 2023, 53(6): 151-158. DOI:10.20051/j.issn.1003-4722.2023.06.020.(in Chinese) |
| [26] |
LI D, YAO A B, CHEN Q F. PSConv: Squeezing feature pyramid into one compact poly-scale convolutional layer[C]//Computer Vision-ECCV 2020. Cham: Springer, 2020. |
| [27] |
LIU W Z, LU H, FU H T, et al. Learning to upsample by learning to sample[C]//2023 IEEE/CVF International Conference on Computer Vision (ICCV). Paris: IEEE, 2024. |
国家自然科学基金(52278507)
山东省重点研发计划(重大科技创新工程)项目(2024CXGC10321)
山东省住房城乡建设科技计划(2024KYKF-JZGYH102)
/
| 〈 |
|
〉 |