Published Paper Details:

This project proposes a deep learning-based system for vehicle detection and counting, leveraging the YOLOv8 model for accurate vehicle detection. The model is trained on a diverse dataset to ensure robustness across various traffic conditions. DeepSORT tracking algorithm is integrated for efficient multi-object tracking, enabling continuous monitoring of vehicle movement. The system processes video feeds in real-time, detecting and counting vehicles as they pass through predefined regions of interest. A Tkinter-based graphical user interface (GUI) is designed for easy input of video files and visualizing the results. The interface also provides real-time updates on detected vehicles and counts. The model is optimized for high accuracy while maintaining performance efficiency. The proposed system is suitable for applications in intelligent transportation systems, including traffic monitoring and automated vehicle counting. The project demonstrates the potential of deep learning techniques in addressing real-world transportation challenges. Future improvements may include handling more complex scenarios and enhancing tracking stability.