The Intelligent Traffic Signal Control System (DEKHO) aims to optimize urban traffic flow using AI-based real-time traffic density analysis. The system dynamically adjusts signal timings based on live vehicle counts and density, ensuring smoother traffic management and reduced congestion at intersections.
- Real-time Traffic Detection: Utilizes computer vision to detect vehicles and calculate traffic density in real time.
- Adaptive Signal Control: Dynamically adjusts traffic signal timings to minimize delays and optimize flow.
- Vehicle Prioritization: Prioritizes emergency vehicles and public transport for efficient traffic management.
- Web-based Dashboard: Provides a Streamlit-based interface for real-time visualization and monitoring.
- Multi-Camera Support: Handles data from multiple intersections for scalable deployment.
| Component | Technology |
|---|---|
| Frontend | Streamlit |
| Backend | FastAPI |
| ML Model | YOLOv8 |
| Database | Firestore |
| Networking | WebSockets |
Clone the DEKHO repository to your local machine:
- Run the application and allow camera access.
- Monitor live traffic density on the Streamlit dashboard.
- Traffic lights adjust dynamically based on detected vehicle count.
- Live Video Input → Captured from a camera at an intersection.
- Vehicle Detection & Counting → YOLOv8 detects cars, bikes, and buses.
- Traffic Density Estimation →
area_counter.pycalculates the percentage. - Signal Adjustment → The backend dynamically modifies timings.
- Data Logging & Analytics → Historical trends stored in Firestore.
- 🚀 Reinforcement Learning (RL) for better traffic predictions.
- 🌍 Edge Computing for real-time processing on IoT devices.
- 📊 Historical Data Insights to improve urban traffic planning.
This project is licensed under the MIT License.
Pull requests are welcome! Feel free to open an issue or suggest improvements.
For inquiries, reach out to [email protected] or visit our GitHub.