ResQ is an end-to-end AI-powered system that detects floods and wildfires in real-time using satellite and drone imagery, and connects disaster victims to rescue teams through a user-friendly web application.
Built with a mission to save lives by reducing response times, ResQ integrates Machine Learning models and a Web App interface — ensuring help reaches victims faster, smarter, and when it’s needed the most.
🌐 Try the Web App: https://resq.lovable.app
- Flood Detection Model: Processes real-time satellite imagery (Sentinel-1 SAR & NOAA-20 VIIRS) to identify flood-affected areas with 92%+ accuracy.
- Victim Detection Model: Detects humans and animals in satellite or drone images and flags high-risk zones for prioritized rescue operations.
- SOS Button: Victims can send a distress signal instantly by pressing a single button.
- Live Location Capture: Captures user's live location and checks for disaster presence via satellite mapping.
- Instant Alerting: Automatically notifies NGOs, rescue teams, and authorities.
- Community Posts: Generates a live post with location and situation to help volunteers and communities act faster.
- Shelter Mapping: Displays nearest shelters and safe zones based on the detected disaster.
- Rescue Tracking: Victims can track the approaching rescue teams through animated map paths.
- Python (for ML Models)
- Streamlit (for ML Model Deployment)
- Lovable.dev (for Web App Development)
- HTML/CSS/TypeScript
- React.js & Tailwind CSS
- Google Maps (for location display)
- Sentinel-1 SAR and NOAA-20 VIIRS Satellite Data
- Roboflow (for Victim Detection Dataset Training)
- Supabase (for backend and database)
| Flood Mapping Example | Wildfire Detection Example |
|---|---|
 |
 |
Watch our full project demo and pitch here:
I am expanding ResQ to detect tsunamis, landslides, and automate real-time social media alerts.
Feel free to fork, star ⭐, and contribute to make disaster rescue smarter and faster!
For collaborations, ideas, or suggestions:
📧 Email: [deekshitha1325@gmail.com]
🔗 LinkedIn: [https://www.linkedin.com/in/deekshitha-m-b02649254/]
"Technology shouldn't wait for disaster. It should act before it's too late."
This project is licensed under the MIT License.