Real-Time Detection of Personal Protective Equipment using YOLOv8
Keywords:
Artificial intelligence, Convolution Neural Network (CNN), Machine learning, Personal protective equipment, RCNN, YOLOv8Abstract
Real-time detection of Personal Protective Equipment (PPE) using YOLOv8 is a critical advancement in enhancing workplace safety, particularly in high-risk environments such as construction sites. This project leverages the capabilities of the YOLOv8 deep learning model to automate the identification and classification of essential safety gear, ensuring compliance with safety regulations. The primary objective of this initiative is to minimize occupational hazards by detecting whether workers are wearing appropriate PPE, including hard hats, masks, and safety vests. The model is trained on a comprehensive dataset that includes images of workers both wearing and not wearing the required equipment, allowing for effective differentiation between compliant and non-compliant scenarios. The YOLOv8 framework is particularly suited for this application due to its real-time object detection capabilities, which enable rapid processing of video feeds from construction sites. This allows for immediate feedback and intervention when safety violations are detected. The system can also be integrated with alert mechanisms to notify supervisors or safety personnel when non-compliance occurs. In testing, the YOLOv8 model demonstrated a commendable accuracy rate of approximately 76.3% after training on a dataset consisting of over 2,800 labeled images. The model was developed using a custom PPE dataset created specifically for this project, which includes 4,500 images and 88,725 labeled instances. Testing showed that the model reached a 97% Mean Average Precision (MAP) and operated at a speed of 25 Frames Per Second (FPS). The outcomes of our project demonstrate that both detection and counting processes performed efficiently, confirming the model's capability for real-time PPE detection on construction sites.
References
R. T. Mohona, S. Nawer, S. Islam, and M. N. Uddin, “A YOLOv8 Approach for Personal Protective Equipment (PPE) Detection to Ensure Workers’ Safety,” 2024 3rd International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE), Apr. 2024, doi: https://doi.org/10.1109/icaeee62219.2024.10561752.
M. Gugssa, L. Li, L. Pu, A. Gurbuz, Y. Luo, and J. Wang, “Enabling near-real-time safety glove detection through edge computing and transfer learning: comparative analysis of edge and cloud computing-based methods,” Engineering Construction & Architectural Management, May 2024, doi: https://doi.org/10.1108/ecam-07-2023-0763.
N. D. Nath, A. H. Behzadan, and S. G. Paal, “Deep learning for site safety: Real-time detection of personal protective equipment,” Automation in Construction, vol. 112, p. 103085, Apr. 2020, doi: https://doi.org/10.1016/j.autcon.2020.103085.
L. Yang, T. Zhao, and L. Li, “Design and Implementation of a Multi-Layer Dynamic Interactive Device Based on YOLO Algorithm and Raspberry Pi-PLC Collaborative Control,” 2025 8th International Conference on Advanced Algorithms and Control Engineering (ICAACE), pp. 2591–2596, Mar. 2025, doi: https://doi.org/10.1109/icaace65325.2025.11020570.
K. Babar, M. U. Yaseen, A. S. Al-Shamayleh, M. Imran, A. H. Al-Ghushami, and A. Akhunzada, “LPN-IDD: A Lightweight Pyramid Network for Image Deraining and Detection,” IEEE Access, vol. 12, pp. 37103–37119, 2024, doi: https://doi.org/10.1109/access.2024.3371484.
X.-Y. Fan, W.-J. Lin, Z. Liu, and L. Zhao, “Reachable set control for nonlinear Markov jump cyber–physical systems with false data injection attacks,” Journal of the Franklin Institute, vol. 361, no. 1, pp. 224–233, Dec. 2023, doi: https://doi.org/10.1016/j.jfranklin.2023.12.013.
G. Gallo, F. D. Rienzo, F. Garzelli, P. Ducange, and C. Vallati, “A Smart System for Personal Protective Equipment Detection in Industrial Environments Based on Deep Learning at the Edge,” IEEE Access, vol. 10, pp. 110862–110878, 2022, doi: https://doi.org/10.1109/ACCESS.2022.3215148.
B. Balakreshnan, G. Richards, G. Nanda, H. Mao, R. Athinarayanan, and J. Zaccaria, “PPE Compliance Detection using Artificial Intelligence in Learning Factories,” Procedia Manufacturing, vol. 45, pp. 277–282, 2020, doi: https://doi.org/10.1016/j.promfg.2020.04.017.
S. Sethi, M. Kathuria, and T. Kaushik, “Face mask detection using deep learning: An approach to reduce risk of Coronavirus spread,” Journal of Biomedical Informatics, vol. 120, p. 103848, Aug. 2021, doi: https://doi.org/10.1016/j.jbi.2021.103848.
W. Z. Khan, E. Ahmed, S. Hakak, I. Yaqoob, and A. Ahmed, “Edge computing: A survey,” Future Generation Computer Systems, vol. 97, pp. 219–235, Aug. 2019, doi: https://doi.org/10.1016/j.future.2019.02.050.
X. Long, W. Cui, and Z. Zheng, “Safety Helmet Wearing Detection Based on Deep Learning,” 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), Mar. 2019, doi: https://doi.org/10.1109/itnec.2019.8729039.
L. López, J. Suárez-Ramírez, M. Alemán-Flores, and N. Monzón, “Automated PPE compliance monitoring in industrial environments using deep learning-based detection and pose estimation,” Automation in Construction, vol. 176, p. 106231, May 2025, doi: https://doi.org/10.1016/j.autcon.2025.106231.
