Virtual Booth

Traffic congestion and red-light violations are significant challenges at urban intersections, creating safety risks and increasing the need for efficient traffic monitoring. Conventional monitoring methods often depend on fixed-time traffic signals and manual surveillance, which may limit real-time traffic analysis and timely violation detection. This project proposes a Smart Intersection Management and Traffic Violation Detection System that integrates Artificial Intelligence (AI), Computer Vision, and edge computing to provide automated and real-time traffic monitoring. The system uses a Raspberry Pi 5 and Camera Module 3 to capture live traffic video, while YOLOv8 performs real-time vehicle detection and OpenCV identifies traffic light conditions. A rule-based approach detects red-light violations by analysing vehicle movement in relation to the traffic signal and stop line. Processed traffic information is transmitted from the Raspberry Pi 5 to a laptop through a 2.4 GHz Wi-Fi connection and stored in a MySQL database. A web-based dashboard provides real-time visualization of vehicle activity, traffic violations, and congestion levels to support traffic monitoring and decision-making. The novelty of the system is its integration of vehicle detection, traffic light recognition, automated red-light violation detection, traffic congestion prediction, database management, and real-time dashboard monitoring within a single platform. The system can benefit traffic authorities, including PDRM, by reducing dependence on manual monitoring and improving situational awareness. Its commercialization potential includes smart-city applications, traffic enforcement, and intelligent transportation management.