GeoShield is an advanced AI-driven platform designed for real-time landslide risk assessment using interactive satellite imagery and deep learning. This project integrates Google Earth Engine for satellite image acquisition and U-Net for precise terrain analysis, providing an intuitive and accessible tool for disaster preparedness. 🌏💡
- 🗺 Interactive Globe Interface: Explore the world and select any location for landslide analysis.
- 🛰 Satellite Imagery Capture: Fetches high-resolution images from Copernicus Sentinel-2.
- 🧠 AI-Powered Predictions: Uses a trained U-Net model to assess landslide risks.
- 🌎 Multi-Map Views: Switch between labeled maps, terrain views, and real-time tracking.
- 📍 Coordinate Input & Auto-Detection: Choose locations via manual input, map selection, or auto-detection.
🖥️ The web-based Streamlit interface provides:
- Map-based location selection
- Real-time satellite image fetching
- Customizable map views for better insights
| Component | Technology Used |
|---|---|
| 🌐 Frontend | Streamlit (Python) |
| 🛰 Satellite Data | Google Earth Engine API |
| 🧠 Deep Learning | U-Net (PyTorch/TensorFlow) |
| 🖼 Image Processing | Custom preprocessing pipeline |
- Select a location on the interactive globe 🌍.
- Capture satellite imagery using Google Earth Engine 🛰️.
- Preprocess the image for model analysis 📊.
- Run the U-Net model to detect landslide risk 🏔️.
- Display prediction results highlighting high-risk zones
⚠️ .
Ensure you have the following installed:
- Python 3.8+
- Streamlit
- Google Earth Engine API
- PyTorch/TensorFlow
- OpenCV & NumPy
# Clone the repository
git clone https://github.com/yourusername/LandslideGuardian.git
cd LandslideGuardian
# Install dependencies
pip install -r requirements.txt
# Run the application
streamlit run app.pyfrom landslide_model import predict_landslide
image = capture_satellite_image(latitude, longitude)
result = predict_landslide(image)
show_result(result)🏆 Our U-Net model has been trained on a large dataset of landslide imagery, achieving high accuracy in predicting landslide risks. The system is continuously improved with updated datasets for better performance.
- 🔍 Integration of AI-based anomaly detection
- 📡 Enhanced satellite data sources
- 📊 Improved visualization techniques
- 🌏 Expansion to flood & earthquake prediction
Want to improve LandslideGuardian? Contributions are welcome! Feel free to:
- ⭐ Star this repository
- 🛠️ Fork & create pull requests
- 📥 Submit feature requests
This project is licensed under the MIT License.
Developed by Team Kitretsu 🎯🚀
For queries, reach out at teamkitretsu@example.com 📩
🌍 Predict & Prevent Landslides with AI 🛡️