Securing Pedestrian Crosswalks in Smart Cities: An Embedded Vision System for Pedestrian Detection and Safety Enhancement

One of the most severe issues in the context of rapidly growing intelligent cities is the safety of pedestrians at urban crosswalks. To address this issue, the system developed an embedded vision system for real-time pedestrian detection and to enhance safety at crosswalks. A camera module captures real-time video, an image processor analyzes the captured data using a pedestrian detection algorithm, and a microcontroller manages the system's other functions. Our system can accurately identify pedestrians inside the crossing area using deep learning-based object detection models. If pedestrians are detected, an LED warning system alerts drivers to either slow down or stop, drastically decreasing the possibility of an accident. A real-time pedestrian count and any applicable safety warnings are also shown on an LCD screen. By connecting with existing systems, this smart technology may significantly improve the security of smart cities for pedestrians. Preliminary testing has demonstrated the technology has great potential for deployment in real-world urban environments.