Intelligent Early Safety Warning Method for High-Altitude Operations on Bridges Based on Personnel Trajectory Prediction

The main construction phase of bridge projects primarily involves high-altitude operations, where the complex site environment and inadequate prior control often lead to safety accidents. An intelligent safety early warning method for high-altitude operations on bridges is designed using deep learning and computer vision techniques, focusing on personnel trajectory prediction. Initially, You Only Look Once (YOLO)v5 is employed to identify the presence of edge guardrails around the high-altitude operation platform, while the Canny edge detection algorithm extracts the platform edges. Dynamic danger zones for high falls are delineated based on the presence or absence of these guardrails. Subsequently, an improved YOLOv5+Deep Simple Online and Realtime Tracking (DeepSORT) multitarget tracking algorithm is used to track construction workers and extract their trajectories. The social long short-term memory (LSTM) algorithm is then utilized to predict the workers’ future movement trajectories. Finally, the ray casting algorithm determines whether these predicted trajectories enter dangerous zones, providing proactive warnings for potential falls. The results demonstrate that this method accurately predicts worker trajectories and provides timely warnings. This approach effectively addresses the limitations of manual supervision in construction safety management, reduces the likelihood of accidents, and offers a novel method for enhancing bridge construction safety management.

Authors

Institutions

Publication Details

Journal
Journal of Construction Engineering and Management
Published
2026-11-19
DOI
https://doi.org/10.1061/jcemd4.coeng-18443
Primary Topic
Occupational Health and Safety Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Intelligent Early Safety Warning Method for High-Altitude Operations on Bridges Based on Personnel Trajectory Prediction

Wei Feng Tian, Reza Andasht Kazeroon, Xingwang Li, Zhiwen Yang et al.
Journal of Construction Engineering and Management
Occupational Health and Safety Research
article

Intelligent Early Safety Warning Method for High-Altitude Operations on Bridges Based on Personnel Trajectory Prediction

Wei Feng Tian, Reza Andasht Kazeroon, Xingwang Li, Zhiwen Yang, Yueliang Wang, Yan Guo, Jianming Wang, Ao Song
article en

Abstract

The main construction phase of bridge projects primarily involves high-altitude operations, where the complex site environment and inadequate prior control often lead to safety accidents. An intelligent safety early warning method for high-altitude operations on bridges is designed using deep learning and computer vision techniques, focusing on personnel trajectory prediction. Initially, You Only Look Once (YOLO)v5 is employed to identify the presence of edge guardrails around the high-altitude operation platform, while the Canny edge detection algorithm extracts the platform edges. Dynamic danger zones for high falls are delineated based on the presence or absence of these guardrails. Subsequently, an improved YOLOv5+Deep Simple Online and Realtime Tracking (DeepSORT) multitarget tracking algorithm is used to track construction workers and extract their trajectories. The social long short-term memory (LSTM) algorithm is then utilized to predict the workers’ future movement trajectories. Finally, the ray casting algorithm determines whether these predicted trajectories enter dangerous zones, providing proactive warnings for potential falls. The results demonstrate that this method accurately predicts worker trajectories and provides timely warnings. This approach effectively addresses the limitations of manual supervision in construction safety management, reduces the likelihood of accidents, and offers a novel method for enhancing bridge construction safety management.

Journal of Construction Engineering and ManagementVol. 153(2)
Xi'an University of Architecture and Technology (CN)
Occupational Health and Safety Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon