Real-Time Pedestrian Safety Enhancement Using Edge AI in Smart Urban Environments

Pedestrian safety in urban areas is a growing concern due to the high density of pedestrian and vehicular traffic, which increases the risk of accidents. Traditional pedestrian safety systems often rely on centralized processing, leading to delays that can compromise safety in critical situations. This paper proposes a novel Edge AI-based system designed to enhance pedestrian safety by leveraging edge computing technologies, allowing real-time detection and response to pedestrian movement. The system can detect pedestrians in real-time and alert drivers through adaptive signaling mechanisms by deploying state-of-the-art object detection models on embedded devices near crosswalks and intersections. The methodology includes a comprehensive evaluation of YOLO-based models optimized for edge deployment, ensuring that both speed and accuracy meet the demanding requirements of urban environments. Experimental results demonstrate the system’s capability to process video inputs at high frame rates with minimal latency, providing a robust solution for improving pedestrian safety in smart cities.