Research on an Automatic Safety Compliance Monitoring System for Construction Sites Based on Computer Vision

View at publisher Summarise with VIRAT
Authors: Hang Chen, Chunlin Liu, Shengfang Qiao, Junzhi Li, Peitao Gao, Hui Liu
Cited by: 0
References: 19

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

To address the problems of low efficiency, strong subjectivity, and poor real-time performance in traditional construction site safety monitoring, this paper proposes an automatic safety compliance monitoring system for construction sites based on computer vision. The system integrates multiple open-source datasets to construct a standardized Personal Protective Equipment (PPE) dataset, compares the performance of YOLOv8-YOLOv12 series models, selects the optimal lightweight YOLOv12 as the core detection model, and develops a Web-based visual management platform. It realizes real-time detection of safety helmet and reflective vest wearing status, violation record management, and hierarchical decision support. Experimental results show that the system achieves a Precision of 96.8 %, Recall of 95.3 %, mAP50 of 96.1 %, and an inference speed of 58 FPS. The Web interface supports data statistics, violation traceability, and intelligent decision suggestions, which can effectively improve the efficiency and compliance level of construction site safety management.

Fields of Study & Subjects

Advanced Neural Network Applications Occupational Health and Safety Research Infrastructure Maintenance and Monitoring

Publication details

Journal / source
International Conference on Electrical Automation and Artificial Intelligence (ICEAAI)
Pages
1253–1257
Publication date
16 January 2026
Language
EN