Work

Automated PPE compliance monitoring in industrial environments using deep learning-based detection and pose estimation

Year: 2025

Type: article

Abstract: In this work, we present a novel AI framework for automated detection of personal protective equipment (PPE) compliance in complex construction and industrial environments. In the context of Industry 4.0, ensuring health and safety standards is vital for protecting workers engaged in construction, repair, or inspection activities. Our framework employs deep learning techniques for worker detection and pose estimation, enabling accurate PPE identification even in challenging conditions. Leveraging the YOLO detection system and the ViTPose model, we tackle significant challenges in PPE monitoring, such as occlusions and long-distance views. A duplicate removal stage further enhances accuracy by correctly assigning PPE items to individual workers. Validated on a diverse dataset covering real-world industrial conditions, our framework demonstrates high precision and recall, particularly for critical PPE like helmets and vests. The experimental results show the framework’s robustness and applicability for continuous safety monitoring and proactive risk management across various industrial settings.

Source: Automation in Construction

DOI: 10.1016/j.autcon.2025.106231

Authors: Leopoldo López, Jonay Suárez-Ramírez, Miguel Alemán-Flores, Nelson Monzón

Institution: Universidad de Las Palmas de Gran Canaria

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

SDGs: Industry, innovation and infrastructure, Decent work and economic growth, Good health and well-being

Open Access status: hybrid

Estimated APC paid: $5,490