Optimizing Traffic Management Through Density-Driven Dynamic Traffic Signaling and Emergency Vehicle Prioritization Using Audio and Video

The unceasing growth of vehicular traffic poses daily challenges, including traffic jams, longer travel times, financial losses, air pollution, and stress. It also hampers emergency services like ambulances, fire trucks, and police vehicles, resulting in the potential harm to both lives and property. Decongesting traffic and prioritizing emergency vehicles are crucial. In this paper, a hybrid solution using dynamic traffic signaling and real-time emergency vehicle detection is proposed. The vehicle detection model calculates vehicular density to control signals dynamically and an audio–video analysis is employed to identify emergency vehicles amidst traffic. The SSD vehicle detection model yields a 97.25% accuracy, while the proposed ensemble model of CNN (audio detection) and DenseNet-169 (video detection) models achieve an accuracy of 98%. Executed on the Intel oneAPI platform, this system significantly improves traffic management efficiency and emergency vehicle detection.