Development of System using Computer Vision to Detect Personal Protective Equipment Violations

This paper introduces a sophisticated computer vision system designed to improve industrial safety by automating the detection of Personal Protective Equipment (PPE) violations. Utilizing the YOLOv8 algorithm, our system not only identifies the absence of essential safety equipment, such as helmets and jackets but also assesses compliance with safety protocols in real-time. The system was rigorously tested across various industrial settings, demonstrating high efficacy with enhanced mean average precision (mAP) compared to existing models. By integrating this system into current safety monitoring frameworks, we offer a dynamic solution that significantly boosts PPE adherence, potentially reducing workplace injuries and aligning with global safety standards.