A Traffic Flow Monitoring System Based on YOLOv8 and DeepSORT for Real-Time Vehicle Tracking

With the rapid increase in vehicle numbers, urban traffic congestion has become a critical challenge, requiring more efficient and adaptive traffic management solutions. Traditional intelligent monitoring systems rely on historical traffic flow data to regulate signal timing, lacking real-time adaptability. To address this limitation, we propose a novel traffic flow monitoring system integrating YOLOv8 and DeepSORT for real-time vehicle tracking. YOLOv8 provides high-precision vehicle detection, while DeepSORT employs advanced tracking algorithms to maintain continuous vehicle trajectories, even in challenging traffic conditions. A key feature of our system is its dynamic vehicle counting function, which not only records the total number of vehicles passing through the detection area but also distinguishes and counts different types of vehicles separately. This functionality enables a more detailed congestion analysis and facilitates data-driven decision-making for adaptive traffic control. By accurately classifying vehicles, tracking their movements, and providing precise vehicle counts for each category, our system significantly enhances the efficiency and scalability of intelligent transportation systems, contributing to more effective and responsive urban traffic management strategies.