2023, 15(3): 20-26. doi: 10.16670/j.cnki.cn11-5823/tu.2023.03.04
基于YOLOv5算法的施工现场不安全状态智能检测
| 1. | 上海工程技术大学化学化工学院，上海 201620 |
| 2. | 武汉科技大学资源与环境工程学院，武汉 430081 |
| 3. | 冶金矿产资源高效利用与造块湖北省重点实验室，武汉 430081 |
Intelligent Detection of Unsafe State on Construction Site Based on Yolov5 Algorithm
| 1. | School of Chemistry and Chemical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China |
| 2. | School of Resources and Environmental Engineering, Wuhan University of Science and Technology, Wuhan 430081, China |
| 3. | Key Laboratory of Hubei Province for Efficient Utilization of Metallurgical Mineral Resources and Block Building, Wuhan 430081, China |
引用本文:
李自强, 任磊, 刘莉, 苗作华. 基于YOLOv5算法的施工现场不安全状态智能检测[J]. 土木建筑工程信息技术,
2023, 15(3): 20-26.
doi: 10.16670/j.cnki.cn11-5823/tu.2023.03.04
Citation:
Ziqiang Li, Lei Ren, Li Liu, Zuohua Miao. Intelligent Detection of Unsafe State on Construction Site Based on Yolov5 Algorithm[J]. Journal of Information Technologyin Civil Engineering and Architecture,
2023, 15(3): 20-26.
doi: 10.16670/j.cnki.cn11-5823/tu.2023.03.04
摘要:为更好地实现施工现场工人的安全监管，利用YOLOv5目标识别算法结合无人机倾斜摄影三维建模技术构建施工现场不安全状态智能检测模型，实现对人、机械等目标的识别与定位。通过实验对比分析确定最优目标识别算法，并构建多目标识别模型，实验结果符合理论猜想，整体识别平均精度达到了91.6%。在识别的基础上借助倾斜摄影三维模型所提供的空间位置信息进一步确定所识别目标的相对位置，从而确定工人的安全状态。这种视觉定位的准确性由三维模型所决定，所以最后通过实验验证了无人机倾斜摄影所构建的三维模型的距离误差在1.5% 左右，范围长度大于35m距离误差将小于1%，从而说明了目标识别模型所识别出物体的距离具有较高的准确性。
Abstract: In order to improve the safety supervision of workers on the construction site, yolov5 target recognition algorithm combined with UAV tilt photography three-dimensional modeling technology are applied to create the intelligent detection model of unsafe state on the construction site to identify and position human, machinery and other targets. Through experimental comparison and analysis, the paper determines the optimal target recognition algorithm and constructs the multi-target recognition model. The results are in line with the theoretical conjecture, and the overall average recognition accuracy reaches 91.6%. After recognizing the target, its relative position is further determined with the spatial position information provided by the tilt photography three-dimensional model, which manages to ensure the safety state of workers. The accuracy of this visual positioning is determined by the three-dimensional model. It is verified that the distance error of the three-dimensional model constructed by UAV tilt photography is about 1.5% and the range length is greater than 35m, hence the distance error will be less than 1%, which notes the high accuracy for the distance of the object recognized by the target recognition model.
| [1] |
姜玥麒, 袁永博, 张明媛. 基于AHP模糊综合评价的建筑工人不安全行为传播性研究[J]. 工程管理学报, 2021, 35(02): 131-136. |
| [2] |
闫高峰. 基于VR技术的建筑工程施工安全管理体系分析[J]. 山西建筑, 2021, 47(09): 189-191.doi: 10.13719/j.cnki.1009-6825.2021.09.069 |
| [3] |
刘文平. 基于BIM与定位技术的施工事故预警机制研究[D]. 清华大学, 2015. |
| [4] |
郭红领, 于言滔, 刘文平, 等. BIM和RFID在施工安全管理中的集成应用研究[J]. 工程管理学报, 2014, 28(04): 87-92. |
| [5] |
赵一秾, 李若熙, 曹语含, 等. 双流卷积网络工人异常行为识别算法研究[J]. 辽宁科技大学学报, 2019, 42(04): 301- 308.doi: 10.13988/j.ustl.2019.04.013 |
| [6] | |
| [7] |
Dong S, He Q, Li H, et al. Automated PPE Misuse Identification and Assessment for Safety Performance Enhancement[C]. 2015, 13(12): 13-16. |
| [8] |
Khairullah M, Habibur Rahman M, M. H B S. BlueAd: A Location based Service using Bluetooth[J]. International Journal of Computer Applications. 2012, 43(15): 19-22.doi: 10.5120/6179-8608 |
| [9] |
Kelm A, Laußat L, Meins-Becker A, et al. Mobile passive Radio Frequency Identification (RFID) portal for automated and rapid control of Personal Protective Equipment (PPE) on sites[J]. Automation in Construction. 2013, 36(36): 38- 52. |
| [10] |
胡劲松, 赵国强, 陈立生, 等. 一种基于RFID安全帽的实时监控系统及其应用方法[P]. 上海: CN103116926A, 2013-05-22. |
| [11] |
Barro-Torres S, Fernández-Caramés T M, Pérez-Iglesias H J, et al. Real-timepersonal protective equipment monitoring system[J]. Computer Communications. 2012, 36(1): 42-50. |
| [12] |
张明媛, 曹志颖, 赵雪峰, 等. 基于深度学习的建筑工人安全帽佩戴识别研究[J]. 安全与环境学报, 2019, 19(02): 535-541. |
| [13] |
高寒, 骆汉宾, 方伟立. 基于机器视觉的施工危险区域侵入行为识别方法[J]. 土木工程与管理学报, 2019, 36(01): 123-128. |
| [14] |
Han S U, Pena-Mora F. Vision-based detection of a unsafe actions of a construction worker: Case study ladder climbing[J]. Journal of Computing in Civil EngIneering, 2013, 27(6): 635-644. |
| [15] |
Kolar Z, Chen H, Luo X. Transfer learning and deep convolutional neural networks for safety guardrail detection in 2D images[J]. Automation in Construction, 2018, 89: 58- 70. |
| [16] |
Fang Q, Li H, Luo X, et al. Detecting non-hardhatuse by a deep learning method from farfield surveillance videos[J]. Automation in Construction, 2018, 85(03): 1-9. |
| [17] |
王伟, 吕山可, 张雨果, 等. 基于BIM与机器视觉技术结合的建筑施工危险区域入侵预警研究[J]. 安全与环境工程, 2020, 27(02): 196-203. |
| [18] |
王毅恒, 许德章. 基于YOLOv3算法的农场环境下奶牛目标识别[J]. 广东石油化工学院学报, 2019, 29(04): 31-35. |
| [19] |
刘晓慧, 叶西宁. 肤色检测和Hu矩在安全帽识别中的应用[J]. 华东理工大学学报(自然科学版). 2014, 40(3): 19-25. |
| [20] |
Rubaiyat A H M, Toma T T, Kalantari-Khandani M, et al. Automatic Detection of Helmet Uses for Construction Safety[C]. 2017, 14(02): 29-32. |
| [21] |
任磊, 苗作华, 李自强, 等. 基于YOLOv3算法的危险区域工人识别[J]. 土木建筑工程信息技术. (网络首发) |
| [22] |
苗作华, 任磊, 王梦婷, 等. 露天矿山无人机倾斜摄影测量数据的准确性探究[J]. 矿业研究与开发, 2022, 42(02): 164-168. |
计量
- PDF下载量(31)
- 文章访问量(1023)
- HTML全文浏览量(490)
