Documents 📘

Note

Feedback and pull requests for this documentation are very welcome.

1. Create a UnifiedAppleVision instance

final vision = UnifiedAppleVision();

2. Set options

vision.executionPriority = VisionExecutionPriority.veryHigh;
vision.analyzeMode = VisionAnalyzeMode.still;

The priority and mode of analysis processing can be set.

Note

About Analysis Processing Modes

There are two modes of analysis processing.

Mode	Description
.still	Suitable for analyzing still images one by one.
.sequential	It is suitable for analyzing a series of images, such as a video. The results of the analysis of the previous image are used for the next analysis. Suitable for object tracking, etc.

3. Start processing

You can analyze images by calling the analyze method.

// create input image
final input = VisionInputImage(
  bytes: image.bytes,
  size: image.size,
);

// analyze
vision.analyze(
  image: input,
  requests: [
    // add requests you wish to perform
    VisionRecognizeTextRequest(
      onResult: (result) {
        final observations = result.ofRecognizeTextRequest; // get casted results
        // some action
      },
      onError: (error) {
        // handle error
      },
    ),
    VisionDetectTextRectanglesRequest(
      onResult: (result) {
        final observations = result.ofDetectTextRectanglesRequest;
        // some action
      },
    ),
  ],
);

For example, if you wish to perform text recognition, add VisionRecognizeTextRequest().