AI-Driven Smart Traffic Signal Control using Computer Vision for Real-Time Traffic Optimization

Rapid urban growth has increased vehicle density, making traditional fixed-timing traffic systems inefficient and unable to handle real-time conditions or prioritize emergency vehicles. In this research AI-based traffic management framework is developed that leverages computer vision techniques to monitor vehicle density and adapt traffic signal timings in real time. The system dynamically adjusts traffic signal timings in response to live traffic conditions, specifically vehicle density at intersections. The system analysis video feeds using a mathematical model to detect and classify vehicles, enabling adaptive signal timing for easy traffic flow and reduced waiting time. A key feature is the automatic detection of emergency vehicles, allowing immediate signal override for faster response. It employs a modular system architecture that includes detection, control, and monitoring modules. Experimental results show improved traffic efficiency and quicker emergency handling, contributing to smarter city traffic management.