https://github.com/vijay-saravanan/advanced-human-life-detection
Portable, real-time embedded system using mmWave radar, microphone, and accelerometer sensor fusion with advanced signal processing and machine learning to detect and locate humans trapped under debris. Features rapid alerts via LCD, LED, buzzer, and is designed for Raspberry Pi deployment in disaster scenarios.
https://github.com/vijay-saravanan/advanced-human-life-detection
disaster-recovery dwt fft landslide machine-learning random-forest-classifier scikit-learn sensor-fusion sensors-data-collection signal-processing vital-signs
Last synced: 15 days ago
JSON representation
Portable, real-time embedded system using mmWave radar, microphone, and accelerometer sensor fusion with advanced signal processing and machine learning to detect and locate humans trapped under debris. Features rapid alerts via LCD, LED, buzzer, and is designed for Raspberry Pi deployment in disaster scenarios.
- Host: GitHub
- URL: https://github.com/vijay-saravanan/advanced-human-life-detection
- Owner: vijay-saravanan
- Created: 2025-10-02T10:45:12.000Z (18 days ago)
- Default Branch: main
- Last Pushed: 2025-10-02T12:06:06.000Z (18 days ago)
- Last Synced: 2025-10-02T12:36:28.787Z (18 days ago)
- Topics: disaster-recovery, dwt, fft, landslide, machine-learning, random-forest-classifier, scikit-learn, sensor-fusion, sensors-data-collection, signal-processing, vital-signs
- Language: Python
- Homepage:
- Size: 13.6 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Advanced-Human-Life-Detection
A portable, real-time rescue tool that fuses mmWave radar, microphone, and accelerometer signals using advanced digital signal processing (FFT, DWT) and machine learning to detect survivors under debris.
![Uploading fullbox.jpeg…]()
## Features
- **Multi-sensor Fusion:** Integrates radar, audio, and vibration data.
- **Signal Processing:** DWT/FFT extraction for robust feature analysis.
- **Machine Learning:** Automatic detection using Random Forest/Custom ML model.
- **Real-time Feedback:** LCD, LED, and buzzer UI for instant alerts.
- **Data Logging:** Stores all measurements and predictions for research.
## Installation
1. Clone this repository:
```
git clone https://github.com/yourusername/human-life-detection.git
cd human-life-detection
```
2. Install dependencies:
```
pip install -r requirements.txt
```
3. Connect Raspberry Pi and sensors (see `/config/config.py`).
## Usage
1. Prepare and connect your hardware.
2. Run:
```
python main.py
```
3. View LCD, LED, and buzzer outputs for detection status.
4. Inspect and analyze sensor data in `ml/sensor_data.csv`.
## File Structure
| File/Folder | Description |
|---------------------|----------------------------------------------|
| `main.py` | Main execution, ML prediction, plotting, UI |
| `sensors/` | Sensor interface classes for radar, mic, acc |
| `processing/` | Signal processing (FFT, DWT), feature extract|
| `ui/` | Hardware UI indicators (LCD, LED, buzzer) |
| `config/` | Hardware pin/address config |
| `utils/` | Data saving utilities |
| `ml/model.pkl` | Trained ML model (RandomForest, etc.) |
| `ml/sensor_data.csv`| Collected sensor data |
## Contributing
Open issues or pull requests to improve hardware integration, add new signal processing features, or adapt model training.
## License
MIT
## Contact
For questions, open an issue or email `vijaysaravanan1609@gmail.com`.