Deep Learning-Based Automatic Helmet Detection System in Construction Site Cameras
Abstract
Ensuring worker safety in high-risk environments such as construction sites is
paramount. Personal protective equipment, particularly helmets, is critical in
preventing severe head injuries. This study aims to develop an automated helmet
detection system using the state-of-the-art YOLOv8 deep learning model to enhance
real-time safety monitoring. The dataset used for the analysis consists of 16,867
images, with various data augmentation and preprocessing techniques applied to
improve the model's robustness. The YOLOv8 model achieved a 96.9% mAP50
score, outperforming other deep learning models in similar studies. The results
demonstrate the effectiveness of the YOLOv8 model for accurate and efficient
helmet detection in construction sites, paving the way for improved safety monitoring
and enforcement in the construction industry.
Collections
DSpace@BEU by Bitlis Eren University Institutional Repository is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 Unported License..