使用谷歌的MLKit在安卓系统上实现一个数字识别器

Question

我正在尝试使用谷歌的安卓MLKit来检测CameraX预览中的数字礼物。

然而，我似乎找不到不基于Kotlin或使用现已弃用的Firebase ML Kit的教程。

我的代码如下所示：

public class MainActivity extends AppCompatActivity {

private final int REQUEST_CODE_PERMISSIONS = 10;
private static final String TAG = "CameraXApp";

private String[] REQUIRED_PERMISSIONS = new String[] {Manifest.permission.CAMERA};

@Override
protected void onCreate(Bundle savedInstanceState) {
    super.onCreate(savedInstanceState);
    setContentView(R.layout.activity_main);

    viewFinder = findViewById(R.id.view_finder);
    // Request camera permissions
    if (allPermissionsGranted()) {
        viewFinder.post(startCamera);
    } else {
        ActivityCompat.requestPermissions(
                this, REQUIRED_PERMISSIONS, REQUEST_CODE_PERMISSIONS);
    }

    // Every time the provided texture view changes, recompute layout
    viewFinder.addOnLayoutChangeListener(new View.OnLayoutChangeListener() {
        @Override
        public void onLayoutChange(
                View v, int left, int top, int right, int bottom, int oldLeft, int oldTop,
                int oldRight, int oldBottom) {
            updateTransform();
        }
    });
}

private TextureView viewFinder;

private final Runnable startCamera = new Runnable() {
    @Override
    public void run() {
        // Create configuration object for the viewfinder use case
        PreviewConfig previewConfig = new PreviewConfig.Builder().build();

        // Build the viewfinder use case
        Preview preview = new Preview(previewConfig);

        // Every time the viewfinder is updated, recompute layout
        preview.setOnPreviewOutputUpdateListener(
                previewOutput -> {
                    // To update the SurfaceTexture, we have to remove it and re-add it
                    ViewGroup parent = (ViewGroup) viewFinder.getParent();
                    parent.removeView(viewFinder);
                    parent.addView(viewFinder, 0);

                    viewFinder.setSurfaceTexture(previewOutput.getSurfaceTexture());
                    updateTransform();
                });

        // Create configuration object for the image capture use case
        ImageCaptureConfig imageCaptureConfig = new ImageCaptureConfig.Builder()
                .setTargetAspectRatio(new Rational(1, 1))
                // We don't set a resolution for image capture; instead, we
                // select a capture mode which will infer the appropriate
                // resolution based on aspect ration and requested mode
                .setCaptureMode(ImageCapture.CaptureMode.MIN_LATENCY)
                .build();

        // Build the image capture use case and attach button click listener
        ImageCapture imageCapture = new ImageCapture(imageCaptureConfig);
        findViewById(R.id.capture_button).setOnClickListener(view -> {
            File file = new File(getExternalMediaDirs()[0], System.currentTimeMillis() + ".jpg");
            imageCapture.takePicture(new ImageCapture.OnImageCapturedListener() {
                @Override
                public void onCaptureSuccess(ImageProxy image, int rotationDegrees) {
                    super.onCaptureSuccess(image, rotationDegrees);

                }
            });
        });

        // Setup image analysis pipeline that computes average pixel luminance
        HandlerThread analyzerThread = new HandlerThread("YourAnalyzer");
        analyzerThread.start();
        ImageAnalysisConfig analyzerConfig =
                new ImageAnalysisConfig.Builder()
                        .setCallbackHandler(new Handler(analyzerThread.getLooper()))
                        // In our analysis, we care more about the latest image than
                        // analyzing *every* image
                        .setImageReaderMode(ImageAnalysis.ImageReaderMode.ACQUIRE_LATEST_IMAGE)
                        .build();

        ImageAnalysis analyzerUseCase = new ImageAnalysis(analyzerConfig);
        analyzerUseCase.setAnalyzer(new YourAnalyzer());

        // Bind use cases to lifecycle
        CameraX.bindToLifecycle((LifecycleOwner) MainActivity.this, preview, imageCapture,
                analyzerUseCase);
    }
};

private void updateTransform() {
    Matrix matrix = new Matrix();

    float centerX = viewFinder.getWidth() / 2f;
    float centerY = viewFinder.getHeight() / 2f;

    // Correct preview output to account for display rotation
    float rotationDegrees;
    switch (viewFinder.getDisplay().getRotation()) {
        case Surface.ROTATION_0:
            rotationDegrees = 0f;
            break;
        case Surface.ROTATION_90:
            rotationDegrees = 90f;
            break;
        case Surface.ROTATION_180:
            rotationDegrees = 180f;
            break;
        case Surface.ROTATION_270:
            rotationDegrees = 270f;
            break;
        default:
            return;
    }

    matrix.postRotate(-rotationDegrees, centerX, centerY);

    // Finally, apply transformations to our TextureView
    viewFinder.setTransform(matrix);
}

/**
 * Process result from permission request dialog box, has the request
 * been granted? If yes, start Camera. Otherwise display a toast
 */
@Override
public void onRequestPermissionsResult(
        int requestCode, String[] permissions, int[] grantResults) {
    if (requestCode == REQUEST_CODE_PERMISSIONS) {
        if (allPermissionsGranted()) {
            viewFinder.post(startCamera);
        } else {
            Toast.makeText(this,
                    "Permissions not granted by the user.",
                    Toast.LENGTH_SHORT).show();
            finish();
        }
    }
}

private boolean allPermissionsGranted() {
    for (String permission : REQUIRED_PERMISSIONS) {
        if (ContextCompat.checkSelfPermission(getBaseContext(), permission) !=
                PackageManager.PERMISSION_GRANTED) {
            return false;
        }
    }
    return true;
}

private static byte[] toByteArray(ByteBuffer buffer) {
    buffer.rewind();
    byte[] data = new byte[buffer.remaining()];
    buffer.get(data);
    return data;
}

private class YourAnalyzer implements ImageAnalysis.Analyzer {
    @Override
    public void analyze(ImageProxy image, int rotationDegrees) {
        Image mediaImage = image.getImage();
        if (mediaImage != null) {
            InputImage inputImage = InputImage.fromMediaImage(mediaImage, rotationDegrees);
            TextRecognizer recognizer = TextRecognition.getClient();
            Task<Text> result =
                    recognizer.process(inputImage)
                            .addOnSuccessListener(new OnSuccessListener<Text>() {
                                @Override
                                public void onSuccess(Text visionText) {
                                    // Task completed successfully
                                    Toast.makeText(MainActivity.this, "SUCCESS", Toast.LENGTH_SHORT).show();
                                    processTextBlock(visionText);
                                }
                            })
                            .addOnFailureListener(
                                    new OnFailureListener() {
                                        @Override
                                        public void onFailure(@NonNull Exception e) {
                                            // Task failed with an exception
                                            //Toast.makeText(MainActivity.this, "FAILED", Toast.LENGTH_SHORT).show();
                                        }
                                    });
        }
    }
}

private void processTextBlock(Text result) {
    // [START mlkit_process_text_block]
    String resultText = result.getText();
    for (Text.TextBlock block : result.getTextBlocks()) {
        String blockText = block.getText();
        Point[] blockCornerPoints = block.getCornerPoints();
        Rect blockFrame = block.getBoundingBox();
        for (Text.Line line : block.getLines()) {
            String lineText = line.getText();
            Point[] lineCornerPoints = line.getCornerPoints();
            Rect lineFrame = line.getBoundingBox();
            for (Text.Element element : line.getElements()) {
                String elementText = element.getText();
                Point[] elementCornerPoints = element.getCornerPoints();
                Rect elementFrame = element.getBoundingBox();
            }
        }
    }
    // [END mlkit_process_text_block]
}

}

不幸的是，OnFailureListener可以完美地工作。但是我从来没有实现过OnSuccessListener来工作。有人知道我错过了什么吗？任何帮助都将不胜感激。1：https://developers.google.com/ml-kit/vision/text-recognition/android#3.-process-the-image 2：https://developer.android.com/training/camerax 3：https://firebase.google.com/docs/ml-kit/recognize-text

Answer 1

下面是ML工具包的CameraX示例：https://github.com/googlesamples/mlkit/blob/master/android/vision-quickstart/app/src/main/java/com/google/mlkit/vision/demo/java/CameraXLivePreviewActivity.java#L411