google_vision 2.0.0+4

Allows you to add Google Visions image labeling, face, logo, and landmark detection, OCR, and detection of explicit content, into cross platform applications.

example/example.md

Google Vision command line example

Provides a basic examples of how the package can be used to perform image processing.

Google Vision object and face detection example

dart run example/object_and_face_detection.dart

import 'package:google_vision/google_vision.dart';

void main() async {
  final googleVision =
      await GoogleVision.withJwt('example/auth.json');

  final painter =
      Painter.fromFilePath('example/young-man-smiling-and-thumbs-up.jpg');

  final cropped = painter.copyCrop(70, 30, 640, 480);

  await cropped.writeAsJpeg('example/cropped.jpg');

  final requests = AnnotationRequests(requests: [
    AnnotationRequest(image: cropped, features: [
      Feature(maxResults: 10, type: 'FACE_DETECTION'),
      Feature(maxResults: 10, type: 'OBJECT_LOCALIZATION')
    ])
  ]);

  print('checking...');

  AnnotatedResponses annotatedResponses =
      await googleVision.annotate(requests: requests);

  print('done.\n');

  for (var annotatedResponse in annotatedResponses.responses) {
    for (var faceAnnotation in annotatedResponse.faceAnnotations) {
      GoogleVision.drawText(
          cropped,
          faceAnnotation.boundingPoly.vertices.first.x + 2,
          faceAnnotation.boundingPoly.vertices.first.y + 2,
          'Face - ${faceAnnotation.detectionConfidence}');

      GoogleVision.drawAnnotations(
          cropped, faceAnnotation.boundingPoly.vertices);
    }
  }

  for (var annotatedResponse in annotatedResponses.responses) {
    annotatedResponse.localizedObjectAnnotations
        .where((localizedObjectAnnotation) =>
            localizedObjectAnnotation.name == 'Person')
        .toList()
        .forEach((localizedObjectAnnotation) {
      GoogleVision.drawText(
          cropped,
          (localizedObjectAnnotation.boundingPoly.normalizedVertices.first.x *
                  cropped.width)
              .toInt(),
          (localizedObjectAnnotation.boundingPoly.normalizedVertices.first.y *
                      cropped.height)
                  .toInt() -
              16,
          'Person - ${localizedObjectAnnotation.score}');

      GoogleVision.drawAnnotationsNormalized(
          cropped, localizedObjectAnnotation.boundingPoly.normalizedVertices);
    });
  }

  await cropped.writeAsJpeg('example/debugImage.jpg');
}

Google Vision text detection example

dart run example/text_detection.dart

final googleVision =
    await GoogleVision.withJwt('example/auth.json');

final painter = Painter.fromFilePath(
    'example/phospholipids_aqueous_solution_structures.png');

final requests = AnnotationRequests(requests: [
  AnnotationRequest(image: painter, features: [
    Feature(maxResults: 10, type: 'TEXT_DETECTION')
  ])
]);

print('checking...');

AnnotatedResponses annotatedResponses =
    await googleVision.annotate(requests: requests);

print('done.\n');

for (var annotatedResponse in annotatedResponses.responses) {
  for (var textAnnotation in annotatedResponse.textAnnotations) {
    GoogleVision.drawText(
        painter,
        textAnnotation.boundingPoly!.vertices.first.x + 2,
        textAnnotation.boundingPoly!.vertices.first.y + 2,
        textAnnotation.description);

    GoogleVision.drawAnnotations(
        painter, textAnnotation.boundingPoly!.vertices);
  }
}

await painter.writeAsJpeg('example/debugImage.jpg');

Google Vision Flutter example

Provides a basic example of how the package can be used to perform image processing from within as a Flutter app

cd example/flutter/vision_demo/
flutter create --platforms ios .
flutter emulators
flutter emulators --launch apple_ios_simulator
flutter devices
flutter run -d [device id]