UAV-Based Automatic System for Seatbelt Compliance Detection at Stop-Controlled Intersections

Transportation agencies often rely on manual surveys to monitor seatbelt compliance; however, these methods are limited by surveyor fatigue, reduced visibility due to tinted windows or low lighting, and restricted coverage to specific locations, making manual surveys prone to errors and unrepresentative of the broader driving population. This paper presents an automated seatbelt detection system leveraging the YOLO11 neural network on video footage from a tethered uncrewed aerial vehicle (UAV). The objectives are to (1) develop a robust system for detecting seatbelt use at stop-controlled intersections, (2) evaluate factors impacting detection accuracy, and (3) demonstrate the potential of UAV-based compliance monitoring. The model was evaluated in real-world applications at a single-lane and a complex multilane stop-controlled intersection in Iowa. Three studies examined key factors influencing detection accuracy: (i) seatbelt-shirt color contrast, (ii) sunlight direction, and (iii) vehicle type. The system’s performance was compared against manual video reviews and large language models (LLMs), with assessments focusing on detection accuracy, resource utilization, and computational efficiency. Overall, the model achieved a mean average precision (mAP) of 0.902, demonstrated high accuracy across the three studies and outperformed manual methods in reliability and efficiency while providing a scalable, cost-effective alternative to LLM-based solutions.