A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Enhancing Construction Site Safety: A Risk Modeling Approach
Traditional risk assessment methods at construction sites typically rely on manual inspections. These assessments are static and primarily conducted before construction, making it challenging to respond in real-time to the constantly changing environment during construction. This paper proposes a real-time risk assessment and dynamic visualization method for construction sites based on multi-source data fusion, combining data obtained through advanced technologies with data and expertise from traditional methods. By collecting and integrating various data sources such as drone images and fixed surveillance videos, the YOLOv8 model is used for target detection, and depth estimation technology is employed to determine the real-time distance between potential hazards and nearby objects. The method incorporates information from hazard source risk assessment reports generated through manual inspections and utilizes an improved TOPSIS method for dynamic risk assessment of the detected results. Finally, the risk distribution is visually presented through heat maps and other visualization techniques. Experimental results demonstrate that this method can efficiently and accurately identify, assess, and warn of risks during the construction process, providing robust support for safety management at construction sites.
Enhancing Construction Site Safety: A Risk Modeling Approach
Traditional risk assessment methods at construction sites typically rely on manual inspections. These assessments are static and primarily conducted before construction, making it challenging to respond in real-time to the constantly changing environment during construction. This paper proposes a real-time risk assessment and dynamic visualization method for construction sites based on multi-source data fusion, combining data obtained through advanced technologies with data and expertise from traditional methods. By collecting and integrating various data sources such as drone images and fixed surveillance videos, the YOLOv8 model is used for target detection, and depth estimation technology is employed to determine the real-time distance between potential hazards and nearby objects. The method incorporates information from hazard source risk assessment reports generated through manual inspections and utilizes an improved TOPSIS method for dynamic risk assessment of the detected results. Finally, the risk distribution is visually presented through heat maps and other visualization techniques. Experimental results demonstrate that this method can efficiently and accurately identify, assess, and warn of risks during the construction process, providing robust support for safety management at construction sites.
Enhancing Construction Site Safety: A Risk Modeling Approach
Advances in Engineering res
Zhao, Gaofeng (editor) / Satyanaga, Alfrendo (editor) / Ramani, Sujatha Evangelin (editor) / Abdel Raheem, Shehata E. (editor) / Li, Chengqian (author) / Li, Manjiang (author) / Xia, Zhengjun (author) / Hou, Xiaoyu (author)
International Symposium on Traffic Transportation and Civil Architecture ; 2024 ; Tianjin, China
2024-09-24
6 pages
Article/Chapter (Book)
Electronic Resource
English
Hardhat-Wearing Detection for Enhancing On-Site Safety of Construction Workers
| Online Contents | 2015
British Library Online Contents | 2002
Online Contents | 2002
Enhancing Motion Trajectory Prediction for Site Safety by Incorporating Attitude toward Risk
| British Library Conference Proceedings | 2017