Yolo Powered Real Time Vehicle Analytics with Predictive ML

Abstract

This paper investigates the use of the YOLOv8 model for real-time vehicle detection. The main objective is to improve both detection accuracy and processing speed, highlighting the efficiency of the YOLOv8 architecture in identifying vehicles from live camera feeds. The model’s performance is evaluated using key metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and overall detection accuracy. The results indicate that YOLOv8 achieves high accuracy along with fast detection speed, making it wellsuited for real-world applications. In particular, it can be effectively applied in systems like adaptive traffic signal control, where accurate and efficient vehicle detection is essential for optimizing traffic flow.