Skip to the main content

Technical gazette, Vol. 32 No. 2, 2025.

Original scientific paper

https://doi.org/10.17559/TV-20241013002055

Automated Detection of Construction Worker Inattention Using IMU Sensors and Visual Focus of Attention

Seungkeon Lee ; Department of AI & Informatics, Graduate School, Sangmyung University, Hongjimun 2-Gil 20, Jongno-Gu, Seoul 03016, Republic of Korea
Meyoung Lee ; Department of AI & Informatics, Graduate School, Sangmyung University, Hongjimun 2-Gil 20, Jongno-Gu, Seoul 03016, Republic of Korea
Hakjin Lee ; Department of Human-Centered Artificial Intelligence, Graduate School, Sangmyung University, Hongjimun 2-Gil 20, Jongno-Gu, Seoul 03016, Republic of Korea
Daesik Jeong ; Division of Software Convergence, Sangmyung University, Hongjimun 2-Gil 20, Jongno-Gu, Seoul 03016, Republic of Korea
Eui Chul Lee ; Department of Human-Centered Artificial Intelligence, Sangmyung University, Hongjimun 2-Gil 20, Jongno-Gu, Seoul 03016, Republic of Korea *

* Corresponding author.

Full text: english pdf 1.246 Kb

page 610-619

downloads: 948

cite

Abstract

Worker inattention is a major contributor to accidents in the construction industry, particularly falls. This study presents a novel method for automatically detecting and monitoring worker inattention in construction training simulations using Inertial Measurement Unit (IMU) sensors and Visual Focus of Attention (VFOA). The proposed system distinguishes between inattention during work tasks and movement, utilizing quaternion data from IMU sensors to infer head pose direction. A custom software program was developed to track inattention in real-time and communicate with work management systems. Validation through simulations with 20 participants demonstrated high correlations (r > 0.93) between predicted and actual measures of inattention. The system accurately detected instances of inattention during both work tasks and movement. This research provides a foundation for enhancing construction safety through automated, real-time inattention monitoring, potentially reducing fall-related accidents in construction environments.

Keywords

fall prevention; human pose; inattention monitoring; inertial measurement unit sensor; quaternion; user datagram protocol

Hrčak ID:

328635

URI

https://hrcak.srce.hr/328635

Publication date:

27.2.2025.

Visits: 1.292 *