A real-time disaster detection and rescue system made with ❤️ using Python, OpenCV & AI models 🤖

DRAMS is an AI-powered system to detect disasters like fire 🔥 and flood 🌊 in real-time. It helps rescue teams respond faster by sending automatic alerts 📢 with snapshots 📸. The project also provides Grad-CAM visualizations 🧠 for model interpretability.

📺 Watch the walkthrough here:

🔗 Click the image or watch on YouTube

• 🔥 Real-time fire detection
• 🌊 Real-time flood detection
• 🖥️ Web streaming interface for live monitoring
• 📸 Automatic alert snapshots on disaster detection
• 🧠 Grad-CAM visualizations for AI predictions
• 🗄️ Stores data & metadata in MongoDB
• 📱 Fully responsive web interface
• ⚡ Quick AI predictions for emergency response

Disaster-Relief-and-Rescue-System/
├── DRAMS/                  # Web app files (Django/Flask)
├── Dataset/                # Fire & flood images/videos
├── firedetector/           # Fire prediction scripts
├── flooddetector/          # Flood prediction scripts
├── alert_snapshots/        # Captured alert images
├── output/                 # Prediction outputs
├── templates/              # HTML templates
├── static/                 # CSS / JS / assets
├── gradcam_visualizer.py   # Grad-CAM visualizer
├── predict_fire.py         # Fire prediction script
├── predict_flood.py        # Flood prediction script
├── train.py                # AI model training script
├── webstreaming.py         # Web streaming interface
├── requirements.txt        # Python dependencies
└── venv/                   # Virtual environment

📸 Real views of DRAMS in action:

📸 Real Outputs & Model Visualizations from DRAMS (Disaster Relief and Management System)

Fire Detection (Grad-CAM)	Flood Detection Outputs	Flood Detection Variants

Model Training Visuals	LR Finder Curve	Color Plot

Raw Output Frame	Grad-CAM (Alt Frame)

📦 Install Dependencies

Copy code
git clone https://github.com/Shristirajpoot/Disaster-Relief-and-Rescue-System.git
cd Disaster-Relief-and-Rescue-System
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

Copy code
python webstreaming.py

🌐 Open http://127.0.0.1:5000 to see live disaster detection

• 🐍 Python 3

• 🌐 Django / Flask

• 🖼️ OpenCV for video processing

• 🤖 TensorFlow / PyTorch for AI models

• 🗄️ MongoDB for dataset storage

• 🎨 HTML5 + CSS3 for web interface

• 📧 Email: [email protected]
• 🔗 GitHub: @Shristirajpoot

This project is licensed under the MIT License.