doc_scan_flutter

doc_scan_flutter is a Flutter package that lets you scan documents as PDF or JPEG using native platform APIs. It automatically detects edges, crops, and enhances colors to give you a clean scan—just like a dedicated scanner app. Inspired by flutter_doc_scanner, but with improved typing and compatibility

what the user scans	the result you get

Demo images courtesy of Google, source. You can expect the same results on iOS.

Installation

Add this to your pubspec.yaml:

dependencies:
  doc_scan_flutter: ^1.0.1

Run:

flutter pub get

Additional setup

iOS

VisionKit is available for iOS 13 and up. To enforce this, edit your Podfile and add at the very top

platform :ios, '13.0'

While you're in your Podfile, also add this at the bottom:

post_install do |installer|
  installer.pods_project.targets.each do |target|
    # You should already have this
    flutter_additional_ios_build_settings(target)

    # add this
    target.build_configurations.each do |config|
      config.build_settings['GCC_PREPROCESSOR_DEFINITIONS'] ||= [
        '$(inherited)',
        'PERMISSION_CAMERA=1',
      ]
    end
  end
end

Finally, in order to request access to the camera, you need to inform the user on why you need this permission, otherwise your app will fail the App Store review. Edit ios/Runner/Info.plist and add:

<key>NSCameraUsageDescription</key>
<string>We need access to your camera to scan documents.</string>

Android

ML Kit and the Google Play Services require a minimum SDK version of 21. Edit your android/app/build.gradle:

android {
    defaultConfig {
        minSdkVersion 21
    }
}

You should then run :

flutter clean
flutter pub get

To make sure everything is clean and up-to-date.

Usage

import 'package:doc_scan_flutter/doc_scan.dart';

try {
    List<String>? result = await DocumentScanner.scan();

    // Scan in PDF
    List<String>? pdfResult = await DocumentScanner.scan(format: DocumentScannerFormat.pdf);


    if (result == null) {
        // the user cancelled
    }
} on DocumentScannerException {
    // Something went wrong, deal with it!
}

Once the scan is complete, you get a list of path to the scanned documents, which will be stored in the platform's temporary folder, which will be purged at some point, so if you need to persist the files locally, make sure to call File('...').rename('...').

License

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