Real Time Traffic Density Measurement using Computer Vision and Dynamic Traffic Control

Abstract

In recent times, traffic jam has become a common problem in the major cities all over the world. In this paper, we propose a dynamic traffic control system by measuring the traffic density at the intersections by real time video feeds and image processing. A video sample was collected and then Mixture of Gaussian algorithm was used for background subtraction method and for foreground detection to keep the count of the cars in each lane. The vehicles are detected by their line of centroid. A movement in centroid confirms a vehicle. The traffic lights at the intersections will change dynamically according to the conditions of traffic that will be detected from the video feeds. In between two intersections, there will be multiple cameras installed to count the number of vehicles entering and leaving each intersection. Furthermore, we restrict the vehicles to take right turns in the intersections. To validate our work, density measurement algorithm, images of live feeds and logics for traffic control are shown in this paper.