The system design phase plays a crucial role in transforming the requirements of a software system into a structured and organized blueprint for implementation. This paper presents the design and implementation of a Smart Traffic Monitoring System that leverages artificial intelligence and deep learning techniques to automate traffic analysis. The proposed system is designed using a modular architecture, ensuring flexibility, scalability, and ease of maintenance. It consists of multiple interconnected modules including user interface, video processing, vehicle detection, lane analysis, and traffic analysis. The system enables users to upload traffic videos through a web-based interface, after which the videos are processed using computer vision techniques. Frames are extracted using OpenCV and analyzed using the YOLO (You Only Look Once) deep learning model for real-time vehicle detection. The system identifies various types of vehicles such as cars, buses, trucks, and motorcycles, and marks them with bounding boxes. Further, it counts vehicles and evaluates traffic density across different lanes to generate meaningful insights.
Hirulkar et al. (Mon,) studied this question.