Journal of Engineering, Project, and Production Management, 2025, 15(4), 2023-81
Combining Computer Vision and Drones for Proactive Construction Site Safety Monitoring
1 Professor,
Department of Construction Engineering, Chaoyang University of
Technology, Taichung, 413310 Taiwan, E-mail: wenderyu@cyut.edu.tw
2 Associate Professor, College of
Civil Engineering, Fuzhou University, China, E-mail: fanbinghui@fzu.edu.cn
(corresponding author).
3 Researcher, Department of
Construction Engineering, Chaoyang University of Technology, Taichung,
413310 Taiwan, E-mail: hkchang@cyut.edu.tw
4 Assistant Professor, Department
of Construction Engineering, Chaoyang University of Technology, Taichung,
413310 Taiwan. E-mail: wdshiau@cyut.edu.tw
5 Research Assistant, Department of
Construction Engineering, Chaoyang University of Technology, Taichung,
413310 Taiwan, E-mail: s10711096@cyut.edu.tw
6 Department of Automation, Robotic
and Mechatronic in Construction, South-Russian State Polytechnic
University, Russia, E-mail: agi.bulgakov@gmail.com
Project Management
Received June 5, 2023; revised June 8, 2024; June 26, 2025; July 4, 2025; accepted July 6, 2025
Available online October 6, 2025
Abstract: Worksite accidents have long been the leading cause of occupational injuries and fatalities worldwide, primarily due to two factors: the open and dynamic nature of the worksite environment and the inadequacy and incompetence of on-site safety managers. Recent advancements in deep learning (DL) and computer vision (CV) offer promising solutions to long-standing challenges in construction safety management. This paper proposes a proactive, real-time monitoring model for construction site safety, inspired by recent research integrating unmanned aerial vehicles (UAVs) with DL-based CV techniques. Specially designed data matrix (DM) tags were affixed to the safety helmets and vests of workers. The model captures DM-tagged images on-site and applies DL-based image recognition algorithms to assess individual risk levels, thereby enabling the implementation of preventive safety measures. Preliminary experimental results show that the model achieved a recall of 97.3% and a precision of 98.3% in worker identification. These findings highlight the practical potential of the proposed approach. The study concludes with a discussion on how the proposed approach could be applied to future advancements in construction safety management.
Keywords: construction safety, UAV, computer vision (CV), proactive site management model.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Yu, W. D., Fan, B. H., Chang, H. K., Hsiao, W. T., Chiang, H. S., and Bulgakov, A. (2025). Combining Computer Vision and Drones for Proactive Construction Site Safety Monitoring. Journal of Engineering, Project, and Production Management, 15(4), 2023-81.
DOI:
10.32738/JEPPM-2023-81
Full Text
