Enhancing the Highway Transportation Systems with Traffic Congestion Detection Using the Quadcopters and CNN Architecture Schema

Traffic congestion is a significant issue in urban areas worldwide, impacting mobility, increasing air pollution, and affecting the well-being of city residents. The Unmanned Aerial Vehicles (UAVs) and object recognition technologies such as the YOLO (You Only Look Once) algorithm are engaging solutions to mitigate traffic congestion. This study focuses on using UAVs as a versatile and cost-effective solution for monitoring highway traffic congestion and enhancing traffic management. UAVs are highly manoeuvrable and can navigate around obstacles to capture detailed imagery of congested areas. The flexibility of UAVs allows drones to access challenging terrain or areas with limited road access, providing valuable insights into traffic conditions from different vantage points. YOLOv8n (YOLOv8 nano) and YOLOv8s (YOLOv8 small) as the lightweight models are utilized for congestion detection with a ratio of 70% for data training and 30% for data validation. This study aims to detect congestion and analyze the traffic flow by combining the latest lightweight YOLO architecture and UAV technology. The results indicate that the combination of the UAV and YOLOv8 model for congestion detection is a promising approach for traffic management.