The rapid growth of vehicles in urban areas has increased the need for intelligent traffic monitoring and automated vehicle analysis systems. Manual vehicle monitoring methods are time-consuming, less efficient, and prone to human errors. To overcome these limitations, this project proposes an AI-Based Vehicle Counting and Crossing Time Analysis System using Computer Vision and Deep Learning technologies. The proposed system uses the YOLOv8 object detection model to detect and track vehicles from traffic video streams in real time. Different types of vehicles such as cars, buses, trucks, and motorcycles are identified automatically using Artificial Intelligence techniques. A virtual checkpoint line is created within the video frame to monitor vehicle movement and count vehicles whenever they cross the predefined region. The system also integrates EasyOCR technology for optional license plate recognition operations. The OCR module extracts alphanumeric characters from vehicle number plates and converts them into machine-readable digital text. The detected vehicle information, crossing activity, and monitoring statistics are displayed through a Streamlit-based dashboard interface.The proposed system improves traffic monitoring efficiency, reduces manual effort, minimizes counting errors, and supports intelligent transportation applications. The project can be used in smart traffic monitoring systems, parking management, highway analysis, toll gate monitoring, and smart city surveillance applications. The developed system demonstrates the successful integration of Artificial Intelligence, Computer Vision, OCR, and Real-Time Video Processing technologies for intelligent vehicle monitoring and analysis.

Keywords – Artificial Intelligence, YOLOv8, Vehicle Detection, Vehicle Counting, OCR, OpenCV, Traffic Monitoring, Computer Vision