Urban traffic congestion is becoming one of the critical problems associated with the increasing population and number of automobiles in cities. Traffic jams not only cause additional delays and stress for drivers, but also increase fuel consumption, transportation costs and air pollution. Even if it seems omnipresent, megacities are the most affected. And its ever-increasing nature makes it imperative to know road traffic density in real time for better signal control and effective traffic management. Traffic controller is one of the critical factors affecting traffic flow. Traffic management systems currently in place are typically static, meaning they do not adjust based on the needs of traffic flow. Our proposed system aims to design a computer vision-based traffic light controller that can adapt to the current traffic situation. It uses live video feed from CCTV cameras at intersections to calculate traffic density in real time by detecting vehicles at the signal and adjusting the green signal time accordingly. Vehicles are categorized as car, bike, bus/truck or rickshaw to get a more accurate estimate of green signal time. We used object detection techniques like YOLO to detect the number of vehicles for each direction. We then set the timers of these traffic lights based on the density of vehicles in each direction and the system thus becomes adaptive. This helps optimize green signal times and traffic is cleared at a much faster rate than a static system, reducing unwanted delays, congestion and waiting time, which in turn will reduce consumption fuel and pollution.

Source code: https://github.com/mihir-m-gandhi/Adaptive-Traffic-Signal-Timer/
Research paper: https://ieeexplore.ieee.org/document/9358334

Email: [email protected]
LinkedIn: https://www.linkedin.com/in/mihir-m-gandhi/

#gpmcoonline project

Please take the opportunity to connect and share this video with your friends and family if you find it useful.