inspiration
Traffic congestion is one of the biggest challenges in urban mobility, especially in rapidly growing cities like New Delhi. Hours are lost daily due to inefficient signal management and unresponsive traffic systems. Our inspiration came from noticing that while cities have installed surveillance cameras everywhere, they are often underutilized mostly serving as passive monitoring tools rather than intelligent, responsive systems. We wanted to turn that untapped visual data into real-time traffic insights.
How we built it We used YOLOv8 for vehicle detection, OpenCV for frame processing, and a Flask-based backend to manage real-time data flow. The system interfaces with simulated traffic signal APIs that adapt timings based on congestion levels. For scalability, we designed it to be deployable on edge devices connected to cloud servers for centralized analytics. Technologies: Python, OpenCV, YOLOv8, Flask, MQTT, and Google Cloud Platform.
Log in or sign up for Devpost to join the conversation.