Abstract
This study addresses construction safety by deploying computer vision techniques, specifically a YOLOv8 model by Ultralytics, to monitor PPE compliance. Targeting helmets, vests, and safety shoes, it aims to mitigate accident risks. The model was trained with 2934 images and validated with 816, achieved a 95% mAP. Emphasizing AI's potential in safety management and occupational health in the construction industry. This research lays groundwork for future AI-based safety enhancements in construction sector, highlighting the industry's pressing need for innovative approaches to reduce occupational hazards and improve compliance standards.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2024 European Conference on Computing in Construction |
| Editors | Marijana Srećković, Mohamad Kassem, Ranjith Soman, Athanasios Chassiakos |
| Number of pages | 8 |
| Publisher | European Council on Computing in Construction |
| Publication date | 2024 |
| Pages | 577-584 |
| ISBN (Electronic) | 978-9-083451-30-5 |
| DOIs | |
| Publication status | Published - 2024 |
| Event | 2024 European Conference on Computing in Construction - Chania, Greece Duration: 14 Jul 2024 → 17 Jul 2024 https://ec-3.org/conference2024/ |
Conference
| Conference | 2024 European Conference on Computing in Construction |
|---|---|
| Country/Territory | Greece |
| City | Chania |
| Period | 14/07/2024 → 17/07/2024 |
| Internet address |
| Series | European Conference on Computing in Construction |
|---|---|
| ISSN | 2684-1150 |