Efficient Traffic Density Calculation for Diverse Road Traffic using Yolov8 Algorithm

Accurate traffic density estimation is vital for effective traffic management. Traditional methods like inductive loop detectors and radar have high costs and scalability issues. This report introduces a method using the YOLO (You Only Look Once) object detection algorithm to calculate traffic density. YOLOv3, fine-tuned with a custom dataset, enhances vehicle detection accuracy. The approach involves collecting video feeds from traffic surveillance cameras, sampling frames at regular intervals, and processing them with YOLOv3 to detect and count the number of vehicles. Traffic density is calculated based on vehicles per unit length of the road. Testing across various lighting conditions and traffic densities showed a detection rate consistently above 95%, validating the method's robustness. YOLO's real-time processing ensures efficient traffic density computation, supporting timely traffic control decisions.EfficientDet algorithm is also detects the vehicles but not accurate as YOLO algorithm as it does not detect the fast moving vehicles which leads to the inefficient use of algorithm.