RM Banana Web UI

A Dockerized web interface for removing invisible AI watermarks from Google Gemini-generated images.

RM Banana provides a clean, modern web UI for the removebanana library. All processing happens locally in your Docker environment with complete privacy.

Upstream Project: https://github.com/denuwanpro/removebanana

Features

• Drag & Drop Interface - Easy image upload with visual feedback
• Privacy First - All processing happens locally in your Docker environment
• Mathematical Precision - Uses inverse alpha blending (no AI guessing)
• Responsive Design - Works on desktop and mobile devices
• NPM Integration - Designed to work with existing Nginx Proxy Manager setups
• Health Checks - Built-in container health monitoring

Architecture

See docs/diagrams/ for editable .drawio source files.

Prerequisites

• Docker & Docker Compose installed
• Existing Nginx Proxy Manager on the shared_net Docker network

Quick Start

1. Clone & Start

git clone https://github.com/yourusername/rmbanana.git
cd rmbanana

# Create shared network if not exists
docker network create shared_net

# Start the container
docker-compose up -d

2. Configure in Nginx Proxy Manager

In your NPM admin panel:

1. Go to Hosts → Proxy Hosts → Add Proxy Host
2. Configure:
   • Domain Names: removebanana.yourdomain.com
   • Scheme: http
   • Forward Hostname / IP: removebanana
   • Forward Port: 3000
3. Save
4. (Optional) Enable SSL under the SSL tab

3. Access the Application

Visit: https://removebanana.yourdomain.com

Usage

1. Upload - Drag and drop or click to select a PNG, JPEG, or WebP image
2. Process - Click "Remove Watermark" to process the image
3. Download - Save the cleaned image to your device

How It Works

The watermark removal uses a mathematical inverse of Google's alpha blending:

Google's watermarking:
  watermarked = α × logo + (1 - α) × original

RemoveBanana reversal:
  original = (watermarked - α × logo) / (1 - α)

No AI or machine learning is involved - it's pure mathematical reconstruction of the original pixels.

Supported AI Image Sources

Source	Support
Google Gemini (all versions)	Full
Imagen 2	Full
Imagen 3	Full
Nano Banana AI	Full

API Reference

Endpoint	Method	Description
/api/health	GET	Health check endpoint
/api/remove-watermark	POST	Upload and process an image
/api/download/:id	GET	Download processed image

Example: Remove Watermark

curl -X POST http://localhost:3000/api/remove-watermark \
  -F "image=@watermarked-image.png"

Response:

{
  "success": true,
  "downloadUrl": "/api/download/abc123",
  "filename": "cleaned-image.png",
  "meta": {
    "width": 1024,
    "height": 1024
  }
}

Example: Health Check

curl http://localhost:3000/api/health

Response:

{
  "status": "healthy",
  "timestamp": "2024-01-15T10:30:00.000Z"
}

Configuration

Environment Variables

Variable	Default	Description
NODE_ENV	production	Node environment
PORT	3000	API server port

Volumes

Path	Description
./backend/uploads	Temporary upload storage
./backend/outputs	Temporary output storage

Network

Uses the external shared_net Docker network (must already exist). The container has no exposed ports on the host - all access goes through your existing Nginx Proxy Manager.

File Structure

rmbanana/
├── backend/
│   ├── package.json      # Node dependencies
│   ├── server.js         # Express API server
│   ├── uploads/          # Temp uploads (auto-cleaned)
│   └── outputs/          # Temp outputs (auto-cleaned)
├── frontend/
│   ├── index.html        # Main HTML
│   ├── styles.css        # Styles
│   ├── app.js            # Frontend logic
│   └── logo.svg          # Logo asset
├── docs/
│   └── diagrams/         # Architecture diagrams
├── Dockerfile            # Container build
├── docker-compose.yml    # Service orchestration
└── README.md             # This file

Development

Local Development

# Install dependencies
cd backend && npm install

# Start the server
npm start

Building the Docker Image

docker build -t rmbanana .

Maintenance

View Logs

# RemoveBanana app logs
docker logs -f removebanana

# All services
docker-compose logs -f

Update

docker-compose pull
docker-compose up -d

Restart

docker-compose restart

Clean Up

# Stop and remove container
docker-compose down

# Remove container and volumes
docker-compose down -v

Troubleshooting

Container won't start

# Check logs
docker logs removebanana

# Ensure shared_net exists
docker network ls

# Rebuild
docker-compose down
docker-compose build --no-cache
docker-compose up -d

Cannot connect from Nginx Proxy Manager

• Verify both containers are on shared_net:

  docker network inspect shared_net

• Check the container is healthy:

  docker ps

• Test from NPM container:

  docker exec <npm_container_name> curl http://removebanana:3000/api/health

Images not processing

• Ensure the image is PNG, JPEG, or WebP format
• Check file size is under 10MB
• Check container logs: docker logs -f removebanana

Security Notes

• The removebanana container has no exposed ports on the host
• All external traffic goes through your existing Nginx Proxy Manager
• Temporary files are auto-deleted after 5 minutes
• No data persists between restarts

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Credits

• Watermark Removal Engine: removebanana by Denuwan Thilakarathna
• Web UI & Dockerization: This project (RM Banana)

License

MIT License - see the LICENSE file for details.

---

Made with care for privacy-conscious users who want control over their AI-generated images.