Traffic Prediction for Intelligent Transportation System using Yolov8 Technique

The efficient management of traffic must be held in any urban area, as it minimizes delays and provides proper timings to emergency response. Machine learning-based solution offers intelligent traffic management at four-way signals for emergency vehicle identification and prioritization. The methodology works in several stages: video data supplied to the system, which is then analyzed via YOLOv8 algorithms to identify emergency vehicles; counting and classification of all vehicles at the intersection; finally analyzing these videos to estimate the optimal time to clear the traffic. With the use of COCO-based video databases, the system accepts input in video format to process the detection of emergency vehicles and deliver alerts to their priority passage in real time. When an emergency vehicle is detected, a prompt appears with the location where the vehicle has been detected and clears the way immediately. When there are no emergency vehicles, the system predicts when the intersection will be cleared based on all vehicles being counted to ease traffic flow. This mirrors the day-to-day traffic management while attending to emergencies, therefore considering congestion reduction and heightened traffic efficiency. The results indicated that intelligent traffic management equipped with advanced methods could greatly enhance urban mobility and emergency response ability.