weebi_barcode_scanner 1.6.5

Self-contained barcode scanner with real-time detection overlay. Features embedded YOLO detection, cross-platform support (Windows/macOS), point-of-sale optimizations, and OpenFoodFacts integration. N [...]

Weebi Barcode Scanner

A Flutter package for barcode scanning (1D & 2D) on laptop(Windows and MacOS) powered by YOLO object detection and ZXing decoding.

This package provides unprecedented support for windows barcode scanning in flutter. The only alternative was focused on QR code through a webview simple_barcode_scanner.

Thanks to computer vision barcode detection (yolo) and adequate image enhancement logic (rust pipeline), decoding results are even more accurate than the FFI zxing integration included in flutter_zxing.

On Android for privacy-concerned scanning prefer barcode_scan2 which wraps zxing java APIs in a more performant way. For non private sensitive use-case go for mobile_scanner which wraps the almighty Google ML Kit barcode.

Features

• Cross-Platform: Windows and macOS support
• AI-Powered Detection: YOLO model for accurate barcode localization
• Multiple Formats: QR codes, Code 128, EAN-13, and more
• Real-Time Processing: Live camera feed with detection overlay
• OpenFoodFacts Integration: Automatic product information lookup

📁 Directory Structure Example

• yolo model is downloaded at class init

your_flutter_app/
├── windows/
│   └── rust_barcode_lib.dll               # 10.87 MB (Windows only)
├── macos/
│   └── Frameworks/
│       └── librust_barcode_lib.dylib      # 22 MB (macOS only)
└── pubspec.yaml

Then run:

flutter pub get

Quick Start

import 'package:weebi_barcode_scanner/weebi_barcode_scanner.dart';

// Scan a barcode with one line of code!
var result = await WeebiBarcodeScanner.scan();

if (result.isSuccess) {
  print('Scanned: ${result.code}');
  print('Format: ${result.format}');
} else if (result.isCancelled) {
  print('User cancelled the scan');
} else if (result.hasError) {
  print('Error: ${result.error}');
}

Basic Usage

import 'package:flutter/material.dart';
import 'package:weebi_barcode_scanner/weebi_barcode_scanner.dart';

class ScannerPage extends StatelessWidget {
  @override
  Widget build(BuildContext context) {
    return Scaffold(
      appBar: AppBar(title: Text('Barcode Scanner')),
      body: BarcodeScannerWidget(
        onBarcodeDetected: (result) {
          print('Scanned: ${result.text}');
          print('Format: ${result.format}');
          
          // Show result dialog
          showDialog(
            context: context,
            builder: (context) => AlertDialog(
              title: Text('Barcode Detected'),
              content: Text('${result.format}: ${result.text}'),
              actions: [
                TextButton(
                  onPressed: () => Navigator.pop(context),
                  child: Text('OK'),
                ),
              ],
            ),
          );
        },
        onError: (error) {
          print('Scanner error: $error');
        },
      ),
    );
  }
}

Point-of-Sale Mode

BarcodeScannerWidget(
  config: ScannerConfig.pointOfSale(),  // Single scan, haptic feedback
  onBarcodeDetected: (result) {
    // Automatically stops scanning after first detection
    Navigator.pop(context, result);
  },
)

Continuous Scanning Mode

BarcodeScannerWidget(
  config: ScannerConfig.continuous(),  // Continuous scanning
  onBarcodeDetected: (result) {
    // Keeps scanning for multiple barcodes
    addToCart(result);
  },
)

🎯 Scanner Configurations

Custom Configuration

ScannerConfig(
  // Detection frequency
  detectionInterval: Duration(milliseconds: 500),
  
  // AI model confidence threshold (0.0-1.0)
  confidenceThreshold: 0.6,
  
  // Image enhancement for damaged barcodes
  enableSuperResolution: true,
  
  // Auto-stop after first detection
  stopAfterFirstScan: true,
  
  // Haptic feedback on detection
  enableHapticFeedback: true,
)

📊 BarcodeResult

class BarcodeResult {
  final String text;                    // Decoded barcode text
  final String format;                  // Barcode format (QR_CODE, EAN_13, etc.)
  final String? productName;            // Product name (via OpenFoodFacts)
  final String? productBrand;           // Product brand
  final Map<String, dynamic>? location; // Barcode location in image
  final double? confidence;             // Detection confidence (0.0-1.0)
  
  bool get hasProductInfo => productName != null;
  bool get hasLocationInfo => location != null;
}

🏪 OpenFoodFacts Integration

Product lookup :

BarcodeScannerWidget(
  onBarcodeDetected: (result) {
    if (result.hasProductInfo) {
      print('Product: ${result.productName}');
      print('Brand: ${result.productBrand}');
      // Additional product data available
    }
  },
)

To enable full product features, add credentials (optional):

# Copy template and add your credentials
cp open_prices_credentials.json.example open_prices_credentials.json
# Edit with your OpenFoodFacts account details

🎨 UI Customization

Split-Screen Layout

BarcodeScannerWidget(
  showProductInfo: true,  // Enables split-screen with product details
  onBarcodeDetected: (result) {
    // Product info automatically displayed on right side
  },
)

Custom Overlay

BarcodeScannerWidget(
  overlayBuilder: (context, detections) {
    return CustomPaint(
      painter: YourCustomOverlayPainter(detections),
    );
  },
)

Debug Information

BarcodeScannerWidget(
  config: ScannerConfig(
    // Enable detailed logging
    enableDebugMode: true,
  ),
  onError: (error) {
    print('Detailed error: $error');
  },
)

🚨 Troubleshooting

Common Issues

1. Camera not working

   • Ensure camera permissions are granted
   • Check that camera is not in use by another app
   • Restart the app if camera appears frozen

2. Poor detection accuracy

   • Ensure good lighting conditions
   • Try adjusting confidenceThreshold (lower = more sensitive)
   • Enable enableSuperResolution for damaged barcodes

3. Performance issues

   • Increase detectionInterval (less frequent detection)
   • Disable enableSuperResolution if not needed

📝 License

MIT License - see LICENSE file for details.

Bundled Components

This package includes several bundled components to provide a seamless integration experience:

1. Weebi YOLO Barcode Detection Model (best.rten)

• File: assets/best.rten
• Source: Hugging Face - weebi/weebi_barcode_detector
• License: AGPL-3.0 (Ultralytics YOLOv8)
• Size: ~12.2MB
• Purpose: Barcode detection AI model for accurate barcode localization

2. Weebi Rust Barcode Library (rust_barcode_lib.dll)

• File: windows/rust_barcode_lib.dll
• Architecture: Windows x64
• License: Proprietary (Weebi.com)
• Size: ~2.1MB
• Purpose: High-performance barcode processing and rxing integration

Features

• YOLO model inference via RTEN runtime
• Image preprocessing and enhancement
• rxing barcode decoding
• Windows-optimized BGRA8888 image handling

3. Dart FFI Bindings

• Files: lib/dart_barcode/
• Purpose: Flutter FFI integration with the Rust library

When using this package:

1. Include attribution in your app credits
2. Respect AGPL-3.0 for the YOLO model

Support

• [e-mail]