flutter_vision 1.1.4

Plugin for managing Yolov5, Yolov8 and Tesseract v5 accessing with TensorFlow Lite 2.x. Support object detection, segmentation and OCR on Android. iOS, Working in progress.

flutter_vision

A Flutter plugin for managing Yolov5, Yolov8 and Tesseract v5 accessing with TensorFlow Lite 2.x. Support object detection, segmentation and OCR on Android. iOS not updated, working in progress.

Installation

Add flutter_vision as a dependency in your pubspec.yaml file.

Android

In android/app/build.gradle, add the following setting in android block.

    android{
        aaptOptions {
            noCompress 'tflite'
            noCompress 'lite'
        }
    }

copied to clipboard

iOS

Comming soon ...

Usage

For YoloV5 and YoloV8 MODEL

1. Create a assets folder and place your labels file and model file in it. In pubspec.yaml add:

  assets:
   - assets/labels.txt
   - assets/yolovx.tflite

copied to clipboard

2. Import the library:

import 'package:flutter_vision/flutter_vision.dart';

copied to clipboard

3. Initialized the flutter_vision library:

 FlutterVision vision = FlutterVision();

copied to clipboard

4. Load the model and labels: modelVersion: yolov5 or yolov8 or yolov8seg

await vision.loadYoloModel(
        labels: 'assets/labelss.txt',
        modelPath: 'assets/yolov5n.tflite',
        modelVersion: "yolov5",
        quantization: false,
        numThreads: 1,
        useGpu: false);

copied to clipboard

For camera live feed

5. Make your first detection: confThreshold work with yolov5 other case it is omited.

Make use of camera plugin

final result = await vision.yoloOnFrame(
        bytesList: cameraImage.planes.map((plane) => plane.bytes).toList(),
        imageHeight: cameraImage.height,
        imageWidth: cameraImage.width,
        iouThreshold: 0.4,
        confThreshold: 0.4,
        classThreshold: 0.5);

copied to clipboard

For static image

5. Make your first detection or segmentation:

final result = await vision.yoloOnImage(
        bytesList: byte,
        imageHeight: image.height,
        imageWidth: image.width,
        iouThreshold: 0.8,
        confThreshold: 0.4,
        classThreshold: 0.7);

copied to clipboard

6. Release resources:

await vision.closeYoloModel();

copied to clipboard

For Tesseract 5.0.0 MODEL

1. Create an assets folder, then create a tessdata directory and tessdata_config.json file and place them into it. Download trained data for tesseract from here and place it into tessdata directory. Then, modifie tessdata_config.json as follow.

{
    "files": [
      "spa.traineddata"
    ]
}

copied to clipboard

2. In pubspec.yaml add:

assets:
    - assets/
    - assets/tessdata/

copied to clipboard

3. Import the library:

import 'package:flutter_vision/flutter_vision.dart';

copied to clipboard

4. Initialized the flutter_vision library:

 FlutterVision vision = FlutterVision();

copied to clipboard

5. Load the model:

await vision.loadTesseractModel(
      args: {
        'psm': '11',
        'oem': '1',
        'preserve_interword_spaces': '1',
      },
      language: 'spa',
    );

copied to clipboard

For static image

6. Get Text from static image:

    final XFile? photo = await picker.pickImage(source: ImageSource.gallery);
    if (photo != null) {
      final result = await vision.tesseractOnImage(bytesList: (await photo.readAsBytes()));
    }

copied to clipboard

7. Release resources:

await vision.closeTesseractModel();

copied to clipboard

About results

For Yolo v5 or v8 in detection task

result is a List<Map<String,dynamic>> where Map have the following keys:

   Map<String, dynamic>:{
    "box": [x1:left, y1:top, x2:right, y2:bottom, class_confidence]
    "tag": String: detected class
   }

copied to clipboard

For YoloV8 in segmentation task

result is a List<Map<String,dynamic>> where Map have the following keys:

   Map<String, dynamic>:{
    "box": [x1:left, y1:top, x2:right, y2:bottom, class_confidence]
    "tag": String: detected class
    "polygons": List<Map<String, double>>: [{x:coordx, y:coordy}]
   }

copied to clipboard

For Tesseract

result is a List<Map<String,dynamic>> where Map have the following keys:

    Map<String, dynamic>:{
      "text": String
      "word_conf": List:int
      "mean_conf": int}

copied to clipboard

Example

Contact

For flutter_vision bug reports and feature requests please visit GitHub Issues