Intelligent Traffic Light Solution for Green and Sustainable Smart City

Abstract

This research aims to develop a smart traffic light system that can improve traffic flow in urban areas. The proposed system uses sensors, cameras, and software to adjust the timing of traffic signals based on real-time traffic conditions. In this study, a Raspberry Pi 4 and MATLAB software was used to build the smart traffic controller. The detection part of the system involves several steps, including removing noise and retrieving information to calculate the number of cars detected. The system then switches traffic lights based on the detected car count. The MATLAB Image Acquisition and Computer Vision toolboxes were used to obtain and analyze the video frames received from the connected cameras. The detector uses the Gaussian Mixture Model to suppress frequently occurring features and to detect abnormal features, which are then used to detect changes caused by moving objects. Morphological operations are used to remove noise from the output. Finally, the system counts the cars detected by the Foreground Detector and switches the traffic lights accordingly. The proposed approach can help reduce traffic congestion and improve the overall flow of traffic in urban areas.