Flutter Blind Watermark

A Flutter plugin for embedding and extracting invisible (blind) watermarks in images using the DWT-DCT-SVD algorithm.

Features

• Invisible Watermarking: Embed watermarks that are imperceptible to the human eye
• Robust: Watermarks survive JPEG compression, cropping, rotation, and other attacks
• High Performance: Native C++ implementation using Eigen library for fast matrix operations
• Cross-platform: Supports both Android and iOS via FFI
• Flexible API: Both synchronous and asynchronous APIs available
• Multiple Watermark Types: Support for text, binary, and image watermarks

How It Works

The algorithm combines three powerful transforms:

1. DWT (Discrete Wavelet Transform): Decomposes the image into frequency sub-bands
2. DCT (Discrete Cosine Transform): Further transforms the coefficients for better embedding
3. SVD (Singular Value Decomposition): Embeds the watermark into singular values for robustness

Installation

Add this to your package's pubspec.yaml file:

dependencies:
  flutter_blind_watermark: ^0.0.2

Then run:

flutter pub get

Quick Start

Embedding a Watermark

import 'package:flutter_blind_watermark/flutter_blind_watermark.dart';

// Simple one-shot API (async, recommended)
final result = await embedStringWatermarkAsync(imageBytes, 'My Secret Watermark');
final watermarkedImage = result.imageBytes;
final wmBitLength = result.wmBitLength; // Save this for extraction!

// Or sync version
final syncResult = embedStringWatermark(imageBytes, 'My Secret Watermark');

Extracting a Watermark

// Async version (recommended)
final extractedText = await extractStringWatermarkAsync(
  watermarkedImageBytes,
  wmBitLength,  // The bit length from embedding
);

// Sync version
final text = extractStringWatermark(watermarkedImageBytes, wmBitLength);

Using BlindWatermark Class

For more control or multiple operations, use the BlindWatermark class:

final bwm = BlindWatermark(
  d1: 36.0,  // Embedding strength (higher = more robust, more artifacts)
  d2: 20.0,
);

try {
  // Async API (recommended for UI apps)
  final result = await bwm.embedStringAsync(imageBytes, 'Hello World', format: 'png');

  // Or sync API
  bwm.readImageBytes(imageBytes);
  bwm.setWatermarkString('Hello World');
  final result = bwm.embedToBytes(format: 'jpg');
} finally {
  bwm.dispose(); // Always dispose when done
}

API Reference

Convenience Functions

Function	Description
embedStringWatermark()	Embed text watermark (sync)
embedStringWatermarkAsync()	Embed text watermark (async)
extractStringWatermark()	Extract text watermark (sync)
extractStringWatermarkAsync()	Extract text watermark (async)

BlindWatermark Class Methods

Embedding

• readImageBytes(Uint8List) - Load image from bytes
• readImageFile(String) - Load image from file path
• setWatermarkString(String) - Set text watermark
• setWatermarkBits(List<bool>) - Set binary watermark
• setWatermarkImageFile(String) - Set image watermark
• embedToBytes({format}) - Embed and return as bytes
• embedToFile(String) - Embed and save to file
• embedStringAsync() - Async embedding with isolate
• embedBitsAsync() - Async bit array embedding

Extraction

• extractStringFromBytes() - Extract text from bytes
• extractStringFromFile() - Extract text from file
• extractBitsFromBytes() - Extract binary from bytes
• extractImageFromBytes() - Extract image watermark
• extractStringAsync() - Async text extraction
• extractBitsAsync() - Async binary extraction
• extractImageAsync() - Async image extraction

Parameters

Parameter	Default	Description
passwordWm	1	Password for watermark encryption
passwordImg	1	Password for image processing
d1	36.0	Embedding strength for first singular value (10-50 recommended)
d2	20.0	Embedding strength for second singular value (5-30 recommended)

Trade-off:

• Higher d1/d2 values = More robust watermark, but more visible artifacts
• Lower d1/d2 values = Better image quality, but less robust watermark

Supported Formats

• Input: PNG, JPEG, WebP, BMP, GIF
• Output: PNG, JPEG, WebP

Example App

Check out the example directory for a complete demo application that shows:

• Embedding text watermarks
• Extracting watermarks
• Comparing sync vs async performance
• Saving watermarked images

Platform Support

Platform	Support
Android	Supported
iOS	Supported
macOS	Coming soon
Windows	Coming soon
Linux	Coming soon
Web	Not supported (requires FFI)

Technical Details

The native implementation is written in C++ and uses:

• Eigen: High-performance linear algebra library
• stb_image: Lightweight image loading/saving
• FFI bindings for Dart integration

License

This project is licensed under the MIT License - see the LICENSE file for details.

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 some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Credits

• Algorithm inspired by blind_watermark by guofei9987
• Uses Eigen library for matrix operations
• Uses stb_image for image I/O

Issues

If you encounter any issues, please file them on the GitHub Issues page.