Advanced Traffic Flow Optimization for Intelligent Transportation System

Even in urban zones, optimizing the vehicular movements is key to reducing waiting times, but more so when we have to think about emergency event response time. This project proposes a machine learning based solution which manages the traffic on four way signals with respect to the emergency vehicles. We use the below methods in this methodology: we check for an emergency vehicle using YOLOv8 and RCNN on video data, we counting and categorize all vehicles at the intersection, and analyze the video data checking the time required to clear the traffic. The system employs a video database based on COCO where th e input is analysed to identify emergency vehicles, issuing appropriate alerts to allow for their passage through the surrounding traffic. If an emergency vehicle is detected, a message shows the position of the vehicle and quickly clears the route. It estimates the time required for clearing the intersection (in the absence of emergency vehicles) based on the total number of vehicles in the system, facilitating traffic without emergency vehicles. The method combines general traffic control and emergency response to decrease congestion and improve traffic efficiency as a whole. Intelligent Traffic Management that utilizes advanced techniques have presented the following results that can greatly enhance urban mobility and emergency response capabilities.