Text Detector

A powerful Flutter plugin for detecting and extracting text from images using native iOS Vision framework. Features an iOS-style UI with advanced selection capabilities and intelligent preprocessing for challenging images.

Features

✨ Core Features

• 📷 Detect text in images using native iOS Vision framework
• 🎯 Drag-to-select multiple text blocks
• 👆 Tap to select/deselect individual text blocks
• 📋 Copy selected text or all detected text
• 🔄 Automatic image rotation correction (EXIF orientation)

🎨 UI Features

• iOS-style selection overlay with smooth animations
• Draggable copy toolbar that can be repositioned
• Visual feedback with boundaries for text blocks
• Pinch to zoom and pan support
• Haptic feedback for interactions

🚀 Advanced Detection

• Intelligent preprocessing for bright/reflective images
• Multi-pass detection for challenging conditions
• Auto-brightness analysis and correction
• Support for 90+ languages with automatic detection
• Small text detection (down to 1% of image height)

Installation

Add text_detector to your pubspec.yaml:

dependencies:
  text_detector: ^1.0.0

iOS Setup

Add the following to your Info.plist:

<key>NSPhotoLibraryUsageDescription</key>
<string>This app needs access to photo library to select images for text detection</string>
<key>NSCameraUsageDescription</key>
<string>This app needs camera access to capture images for text detection</string>

Quick Start

Basic Usage

import 'package:text_detector/text_detector.dart';

// Create detector instance
final textDetector = TextDetector();

// Detect text in an image
final textBlocks = await textDetector.detectText(
  imagePath: '/path/to/image.jpg',
);

// Process detected text
for (final block in textBlocks) {
  print('Text: ${block.text}');
  print('Confidence: ${block.confidence}');
  print('Position: x=${block.x}, y=${block.y}');
}

Using the Pre-built Widget

The easiest way to add text detection to your app:

import 'package:text_detector/text_detector.dart';

TextDetectorWidget(
  imagePath: imagePath,
  onTextCopied: (text) {
    print('Copied: $text');
  },
  onTextBlocksSelected: (blocks) {
    print('Selected ${blocks.length} blocks');
  },
)

Advanced Configuration

// Detect with custom settings
final textBlocks = await textDetector.detectText(
  imagePath: imagePath,
  recognitionLevel: RecognitionLevel.accurate, // or .fast
  languages: ['en-US', 'es-ES'], // Specify languages
  enhanceForBrightness: true, // Auto-enhance bright images
  preprocessingLevel: 'auto', // auto/none/light/moderate/aggressive
  multiPass: true, // Multiple detection passes
);

Widget Customization

TextDetectorWidget(
  imagePath: imagePath,
  autoDetect: true, // Auto-detect on load
  backgroundColor: Colors.black,
  showUnselectedBoundaries: true, // Show grey boundaries
  loadingWidget: CustomLoadingIndicator(),
  onTextCopied: (text) {
    // Handle copied text
  },
)

Example App

Check the /example folder for a complete implementation with:

• Image picker integration
• Camera support
• iOS-style UI
• Selection and copy functionality

Run the Example

cd example
flutter run

API Reference

TextDetector

Main class for text detection:

Future<List<TextBlock>> detectText({
  required String imagePath,
  RecognitionLevel recognitionLevel = RecognitionLevel.accurate,
  List<String>? languages,
  bool enhanceForBrightness = true,
  String preprocessingLevel = 'auto',
  bool multiPass = true,
})

TextBlock

Detected text information:

class TextBlock {
  final String text;
  final double confidence;
  final double x, y;
  final double width, height;
}

TextDetectorWidget

Ready-to-use widget with selection UI:

TextDetectorWidget({
  required String imagePath,
  Function(String)? onTextCopied,
  Function(List<TextBlock>)? onTextBlocksSelected,
  bool autoDetect = true,
  RecognitionLevel recognitionLevel = RecognitionLevel.accurate,
  Widget? loadingWidget,
  Color backgroundColor = Colors.black,
  bool showUnselectedBoundaries = true,
})

Performance Tips

1. For better accuracy: Use RecognitionLevel.accurate (default)
2. For faster detection: Use RecognitionLevel.fast
3. For bright images: Keep enhanceForBrightness: true (default)
4. For normal images: Set multiPass: false for faster processing

Handling Challenging Images

The plugin automatically handles:

• 📸 Rotated images (portrait/landscape/upside-down)
• ☀️ Overexposed or bright images
• 🔦 Images with reflections or glare
• 📝 Low-contrast text
• 🔤 Small text

Platform Support

• ✅ iOS 13.0+ (uses Vision framework with VNRecognizeTextRequest)
• 🚧 Android (coming soon with ML Kit)

Contributing

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

License

MIT License - see LICENSE file for details

Author

Created with ❤️ using Flutter and native iOS Vision framework