Please use this identifier to cite or link to this item: https://accedacris.ulpgc.es/handle/10553/138101
Title: Automated PPE compliance monitoring in industrial environments using deep learning-based detection and pose estimation
Authors: López, Leopoldo
Suárez Ramírez, Jonay 
Alemán Flores, Miguel 
Monzón López, Nelson Manuel 
UNESCO Clasification: 120304 Inteligencia artificial
Keywords: PPE compliance
Deep learning
Worker detection
Pose estimation
Safety and environment (HSE)
Issue Date: 2025
Project: Detección precisa mediante Inteligencia Artificial deeventos de interés en escenas de playa, costa y litoral. 
Journal: Automation in Construction 
Abstract: This paper presents an AI framework for automated detection of personal protective equipment (PPE) compliance in complex construction and industrial environments. Ensuring health and safety standards is essential for protecting workers engaged in construction, repair, or inspection activities. The framework leverages deep learning techniques for worker detection and pose estimation to enable accurate PPE identification under challenging conditions. The framework components are replaceable, and employ the InternImage-L detector for worker detection, ViTPose for pose estimation, and YOLOv7 for PPE recognition. A duplicate removal stage, combined with pose information, ensures PPE items are accurately assigned to individual workers. The approach addresses challenges like shadows, partial occlusions, or densely grouped workers. Evaluated on diverse datasets from real-world industrial settings, the framework achieves competitive precision and recall, particularly for critical PPE like helmets and vests, demonstrating robustness for safety monitoring and proactive risk management.
URI: https://accedacris.ulpgc.es/handle/10553/138101
ISSN: 0926-5805
DOI: 10.1016/j.autcon.2025.106231
Source: Automation in Construction[ISSN 0926-5805],v. 176, (Agosto 2025)
Appears in Collections:Artículos
Adobe PDF (10,29 MB)
Show full item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.