Context-Aware PPE Detection for Construction Site Safety Using Enhanced YOLOv11

Abstract

Construction sites are among the most hazardous work environments, where workers face risks from falling objects, heavy machinery, electrical hazards, and exposure to dangerous materials. Ensuring proper use of Personal Protective Equipment (PPE) such as helmets, safety vests, gloves, boots, and masks is critical to minimizing accidents and injuries. Traditional PPE compliance monitoring relies heavily on manual supervision, which is prone to human error, delayed responses, and limited coverage, especially on large-scale construction sites. To overcome these challenges, this project proposes a context-aware PPE detection system for construction site safety using the enhanced YOLOv11 deep learning model. The system is capable of real-time detection and classification of multiple PPE categories, even in complex site conditions involving occlusions, varying lighting, and crowded scenes. By automating monitoring, the framework ensures continuous vigilance, significantly reducing the dependence on manual checks and enhancing workplace safety standards. The proposed system not only identifies compliance violations instantly but also provides immediate alerts through SMS and mobile notifications, allowing supervisors to take corrective actions without delay. Integration with real-time video feeds enables proactive risk management, data-driven insights, and improved regulatory compliance for site managers. Moreover, the system’s advanced object detection capabilities facilitate accurate monitoring of multiple workers simultaneously, ensuring consistent adherence to safety protocols across the construction site. By combining real-time surveillance with automated notifications, this solution fosters a culture of safety consciousness, minimizes accidents due to negligence, and supports efficient safety management, ultimately creating a safer and more compliant work environment for all personnel on site.