An Edge-AI Framework for Real Time PPE Compliance Monitoring at Construction Sites Using IOT Device

Construction industry is one of the fastestgrowing sectors in India and a major contributor to employment and infrastructure development. The design focuses on three operational states for each PPE item: Wearing, Partially Wearing, and Not Wearing. This distinction is important because a simple binary detector can fail to capture unsafe conditions such as an incorrectly positioned helmet or incomplete use of protective equipment. However, it is also one of the most hazardous industries due to the involvement of heavy machinery, work at height, electrical operations, and manual handling of materials. High occupational injury and fatality rate in the construction industry is a serious global concern. Construction accidents lead to loss of life, injuries, project delays, and financial losses. Research has shown that deep-learning based vision systems can automatically assess hard-hat and safety-vest compliance from images and video, while dedicated PPE detectors have demonstrated the feasibility of detecting multiple PPE categories in construction environments.