Real-Time Vehicle Detection Model Using YOLOv8

Abstract

ABSTRACT: With the rapid advancement of intelligent transportation systems, real-time vehicle detection has become a critical component in traffic monitoring, autonomous driving, and road safety applications. This research explores the application of the YOLO (You Only Look Once) deep learning model for fast and accurate vehicle detection. By leveraging a convolutional neural network-based architecture, the proposed system efficiently detects and classifies vehicles in real-world traffic scenarios. The model is trained on a diverse dataset, incorporating various environmental conditions to enhance robustness. Experimental results demonstrate that the proposed YOLO-based approach achieves high detection accuracy while maintaining low computational cost, making it suitable for real-time deployment in smart traffic management systems. To further enhance detection accuracy, this study integrates advanced data augmentation techniques and optimization strategies, including transfer learning and anchor box refinement. The findings suggest that YOLO-based vehicle detection can significantly improve traffic flow analysis, automated surveillance, and accident prevention systems. Future research will focus on integrating sensor fusion techniques and improving small-object detection for more comprehensive vehicle recognition. Keywords: Vehicle Detection, YOLOv8, Real-Time Detection, Machine Learning, Traffic Monitoring, Autonomous Driving.