micro-doppler-web-app

This repository hosts a web application for micro-Doppler-based target classification, focusing on differentiating between drones and birds using radar signatures. The model is built using a fine-tuned VGG19 architecture for classification.

Tech Stack

• Frontend: HTML, CSS, JavaScript.
• Backend: Flask.
• Machine Learning: Python (PyTorch, Scikit-learn).
• Visualization: Matplotlib.
• Deployment:
  • Containerization: Docker.
  • Application Server: Gunicorn.
  • Reverse Proxy: Nginx.
  • Cloud Platform: AWS EC2.

Project Structure

micro-doppler-web-app/
├── static/                    # Static files like css, js, images, etc
├── templates/                 # Frontend html files as templates for Flask
├── app.py                     # Main Flask App
├── venv                       # virtual env Folders
├── .gitignore
├── monitor.py                 # file to monitor errors and logs
├── requirements.txt           # flask requirements
├── .env                       # General environment variables
└── README.md                  # Backend documentation