flutter_mrz_scanner_enhanced 3.2.0

Scans MRZ (Machine Readable Zone) from identity documents for iOS and Android.

Flutter MRZ Scanner Enhanced

Based on

A community-maintained fork of the original flutter_mrz_scanner package with significant improvements to MRZ scanning reliability and camera UX.

✨ Key Enhancements

• Improved text recognition accuracy through advanced image preprocessing
• Optimized camera overlay UI for better user experience
• Enhanced image processing pipelines
• Modernized dependencies and null-safety support
• Improved error handling and validation
• Better platform compatibility

🚧 Active Development

As I'm actively using this in production, expect regular updates including:

Performance Improvements 🚀

• Flutter Isolate Integration: Offload heavy image processing and OCR tasks to background isolates for smoother UI performance.
• Real-Time MRZ Detection: Implement live feedback for MRZ detection with visual indicators and dynamic UI updates.

Feature Enhancements ✨

• Machine Learning Model Optimizations: Improve OCR accuracy with updated ML models and preprocessing pipelines.
• Customizable UI Components: Allow developers to fully customize the camera overlay, scanning UI, and feedback animations.

Stability & Maintenance 🔧

• Improved Error Handling: Better error reporting and recovery mechanisms for edge cases.
• Cross-Platform Compatibility: Ensure consistent behavior across iOS, Android, and web platforms.
• Community-Driven Features: Prioritize features and fixes based on community feedback.

🙏 Acknowledgments

This package builds upon the work of:

• @olexale (Oleksandr Leushchenko) - Original creator
• @makhosi6 (Makhosandile) - Early contributor
• @eusopht2021 - Community contributor

Contributions welcome! Please report issues and feature requests on GitHub.

🚨 Project Status & Known Issues

Notice: I will not be pushing future updates to this package due to my inability to test on iOS devices and time constraints.

Common Issues

• OCR accuracy is generally good, but sometimes parsing fails even when the OCR result is correct. This may be an issue with the MRZ Parser. (You can see the raw OCR output in the example code, as I have added print statements before parsing.)
• Fotoapparat dependency: The package needs to switch to a working fork of Fotoapparat. Maven issues need to be fixed, as Fotoapparat is now published to Maven Central, not JitPack.
• Cropped MRZ image display: The package still needs to properly display the cropped MRZ area image.

---

Supported formats:

• TD1
• TD2
• TD3
• MRV-A
• MRV-B

Usage

Import the package

Add to pubspec.yaml

dependencies:
  flutter_mrz_scanner_enhanced: <latest_version_here>

For iOS

Set iOS deployment target to 12. The plugin uses the device camera, so do not forget to provide the NSCameraUsageDescription. You may specify it in Info.plist like that:

    <key>NSCameraUsageDescription</key>
    <string>SCANNING MRZ REQUIRE CAMERA PERMISSIONS</string>

For Android

Add

<uses-permission android:name="android.permission.CAMERA" />

to AndroidManifest.xml

Use the widget

Use MRZScanner widget:

MRZScanner(
  withOverlay: true, // Mandatory for proper document cropping
  onControllerCreated: (controller) =>
    onControllerCreated(controller),
  )

Refer to example project for the complete app sample.

Acknowledgements

• Anna Domashych for helping with mrz_parser implementation in Dart
• Anton Derevyanko for hours of Android-related discussions
• Mattijah for beautiful QKMRZScanner library

License

flutter_mrz_scanner_enhanced is released under a MIT License. See LICENSE for details.