|
基于改进的YOLOv8n模型桥梁裂缝检测识别算法
|
Abstract:
本文针对我国桥梁数量众多、维护需求迫切的背景,提出了一种改进的YOLOv8n模型,以提升桥梁裂缝检测的准确性和效率。通过引入全局注意力机制(GAM),研究成功增强了模型从复杂背景中提取显著特征的能力,结合Wise-IOU损失函数,优化了对难以检测样本的关注度。实验结果表明,改进后的YOLOv8n在精确度、平均精度均值(mAP)和F1分数上较原始模型分别提升了2.4、1.1和1.3个百分点,尽管召回率略有下降,但整体性能仍优于当前主流检测模型。本文的贡献在于通过技术创新,提高了桥梁裂缝的自动化检测能力,填补了传统人工检测方法的效率和主观性缺陷,为桥梁维护提供了更为高效、可靠的解决方案。未来,模型的进一步优化及在更广泛的实际应用中的验证,将是提升桥梁安全性和耐久性的关键。
This study addresses the urgent maintenance needs of the numerous bridges in China by proposing an improved YOLOv8n model to enhance the accuracy and efficiency of bridge crack detection. By incorporating a Global Attention Mechanism (GAM), the research successfully enhances the model’s ability to extract salient features from complex backgrounds. Additionally, the Wise-IOU loss function optimizes the model’s focus on hard-to-detect samples. Experimental results indicate that the improved YOLOv8n achieves increases of 2.4, 1.1, and 1.3 percentage points in accuracy, mean average precision (mAP), and F1 score, respectively, compared to the original model. Although the recall rate slightly declined, the overall performance remains superior to current mainstream detection models. The contribution of this research lies in the technological innovation that improves the automation capability of bridge crack detection, addressing the efficiency and subjectivity deficiencies of traditional manual inspection methods, thus providing a more efficient and reliable solution for bridge maintenance. Future work will focus on further optimizing the model and validating it in a broader range of practical applications, which will be key to enhancing bridge safety and durability.
[1] | 《中国公路学报》编辑部. 中国桥梁工程学术研究综述∙2021 [J]. 中国公路学报, 2021, 34(2): 1-97. |
[2] | 交通运输部. 2022年交通运输行业发展统计公报[N]. 中国交通报, 2023-06-16(002). |
[3] | 梁栋, 李英俊, 张少杰. 融合改进ResNet-14和RS-Unet模型的混凝土桥梁裂缝识别[J]. 北京交通大学学报, 2023, 47(3): 10-18. |
[4] | 岳清瑞, 幸坤涛, 惠云玲. GB 50144《工业建筑可靠性鉴定标准》(报批稿)的修订及主要内容[J]. 工业建筑, 2016, 46(11): 152-156, 167. |
[5] | 刘宇飞, 齐玉, 李保罗, 等. 工程结构运维多场景的敏感服役指标智能识别[J]. 建筑结构学报, 2022, 43(10): 1-15. |
[6] | 陈榕峰, 徐群丽, 秦凯强. 桥梁裂缝智能检测系统的研究新进展[J]. 公路, 2019, 64(5): 101-105. |
[7] | 丁威, 夏哲, 舒江鹏, 等. 基于负压吸附爬壁机器人和Transformer的混凝土桥塔裂缝识别检测[J]. 中国公路学报, 2024, 37(2): 53-64. |
[8] | 苏卫国, 王景霄. 基于YOLOv3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. |
[9] | 岳清瑞, 徐刚, 刘晓刚. 桥梁裂缝智能识别与监测方法研究[J]. 中国公路学报, 2024, 37(2): 16-28. |
[10] | 彭雨诺, 刘敏, 万智, 等. 基于改进YOLO的双网络桥梁表观病害快速检测算法[J]. 自动化学报, 2022, 48(4): 1018-1032. |
[11] | 杨国俊, 齐亚辉, 石秀名. 基于数字图像技术的桥梁裂缝检测综述[J]. 吉林大学学报(工学版), 2024, 54(2): 313-332. |
[12] | Majidifard, H., Jin, P., Adu-Gyamfi, Y. and Buttlar, W.G. (2020) Pavement Image Datasets: A New Benchmark Dataset to Classify and Densify Pavement Distresses. Transportation Research Record: Journal of the Transportation Research Board, 2674, 328-339. https://doi.org/10.1177/0361198120907283 |
[13] | Zhang, C., Chang, C. and Jamshidi, M. (2019) Concrete Bridge Surface Damage Detection Using a Single‐Stage Detector. Computer-Aided Civil and Infrastructure Engineering, 35, 389-409. https://doi.org/10.1111/mice.12500 |
[14] | Zhang, Y., Huang, J. and Cai, F. (2020) On Bridge Surface Crack Detection Based on an Improved YOLO V3 Algorithm. IFAC-PapersOnLine, 53, 8205-8210. https://doi.org/10.1016/j.ifacol.2020.12.1994 |
[15] | 邹俊志, 杨建喜, 李昊, 等. 复杂背景下基于改进YOLOv3算法的桥梁表观病害识别[J]. 铁道科学与工程学报, 2021, 18(12): 3257-3266. |
[16] | Zhang, L., Yang, F., Daniel Zhang, Y. and Zhu, Y.J. (2016) Road Crack Detection Using Deep Convolutional Neural Network. 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, 25-28 September 2016, 3708-3712. https://doi.org/10.1109/icip.2016.7533052 |
[17] | Fan, R., Bocus, M.J., Zhu, Y., Jiao, J., Wang, L., Ma, F., et al. (2019) Road Crack Detection Using Deep Convolutional Neural Network and Adaptive Thresholding. 2019 IEEE Intelligent Vehicles Symposium (IV), Paris, 9-12 June 2019, 474-479. https://doi.org/10.1109/ivs.2019.8814000 |
[18] | 柏嘉洛. 基于深度学习的端到端道路裂缝检测技术研究[D]: [硕士学位论文]. 武汉: 华中科技大学, 2020. |
[19] | 冯卉. 基于深度学习的道路裂缝识别算法研究与实现[D]: [硕士学位论文]. 北京: 北京邮电大学, 2019. |
[20] | Peng, X., Zhong, X., Zhao, C., Chen, A. and Zhang, T. (2021) A UAV-Based Machine Vision Method for Bridge Crack Recognition and Width Quantification through Hybrid Feature Learning. Construction and Building Materials, 299, Article ID: 123896. https://doi.org/10.1016/j.conbuildmat.2021.123896 |
[21] | Yang, J., Fu, Q. and Nie, M. (2020) Road Crack Detection Using Deep Neural Network with Receptive Field Block. IOP Conference Series: Materials Science and Engineering, 782, Article ID: 042033. https://doi.org/10.1088/1757-899x/782/4/042033 |
[22] | Xiong, C., Zayed, T. and Abdelkader, E.M. (2024) A Novel YOLOv8-Gam-Wise-IoU Model for Automated Detection of Bridge Surface Cracks. Construction and Building Materials, 414, Article ID: 135025. https://doi.org/10.1016/j.conbuildmat.2024.135025 |