Question

我正在为我的Android OCR应用程序使用Tesseract库，然后我需要为每个角色获得边界框，所以我按照本教程，但是当我编写此代码时，它显示错误，这是我的代码：

TessBaseAPI baseApi = new TessBaseAPI();
baseApi.setDebug(true);
baseApi.init(Path, Lang);
baseApi.setImage(ReadFile.readBitmap(BitmapBiner));
baseApi.setVariable(TessBaseAPI.VAR_CHAR_BLACKLIST, CharacterBlacklist);
baseApi.setVariable(TessBaseAPI.VAR_CHAR_WHITELIST, CharacterWhitelist);
String RecognizedText = baseApi.getUTF8Text();
List<Rect> characterBoundingBoxes = baseApi.getCharacters().getBoxRects();

      BitmapBiner = BitmapBiner.copy(Bitmap.Config.RGB_565, true);
      Canvas canvas = new Canvas(BitmapBiner);

      // draw bounding box for each character
      for (int i = 0; i < characterBoundingBoxes.size(); i++) {
          paint.setAlpha(0xFF);
          paint.setColor(0xFF00CCFF);
          paint.setStyle(Style.STROKE);
          paint.setStrokeWidth(1);
          Rect r = characterBoundingBoxes.get(i);
          canvas.drawRect(r, paint);
      }

然后，它显示第8行的错误，它说“方法getCharacters（）未定义类型TessBaseAPI”。所以我决定使用另一种方式，它是ResultIterator，这是我的代码：

    TessBaseAPI baseApi = new TessBaseAPI();
    baseApi.setDebug(true);
    baseApi.init(Path, Lang);
    baseApi.setImage(ReadFile.readBitmap(BitmapBiner));
    baseApi.setVariable(TessBaseAPI.VAR_CHAR_BLACKLIST, CharacterBlacklist);
    baseApi.setVariable(TessBaseAPI.VAR_CHAR_WHITELIST, CharacterWhitelist);
    String RecognizedText = baseApi.getUTF8Text();  

    final ResultIterator iterator = baseApi.getResultIterator();
    String lastUTF8Text;
    float lastConfidence;
    int[] lastBoundingBox;
    int count = 0;
    iterator.begin();
    do {
        lastUTF8Text = iterator.getUTF8Text(PageIteratorLevel.RIL_SYMBOL);
        lastConfidence = iterator.confidence(PageIteratorLevel.RIL_SYMBOL);
        lastBoundingBox = iterator.getBoundingBox(PageIteratorLevel.RIL_SYMBOL);
        count++;
    } while (iterator.next(PageIteratorLevel.RIL_SYMBOL));

      BitmapBiner = BitmapBiner.copy(Bitmap.Config.RGB_565, true);
      Canvas canvas = new Canvas(BitmapBiner);

      // draw bounding box for each character
      for (int i = 0; i < lastBoundingBox.length; i++) {
        paint.setAlpha(0xA0);
        paint.setColor(Color.RED);
        paint.setStyle(Style.STROKE);
        paint.setStrokeWidth(1);
        Rect r = new Rect(lastBoundingBox[0], lastBoundingBox[1],                       
                          lastBoundingBox[2], lastBoundingBox[3]);
        canvas.drawRect(r, paint);
      }

到目前为止它工作得很好，但现在它现在只有最后一个字符得到了边界框，例如word是“DOG”，...所以在那张图片中唯一的字符是边界框是“G”,. ......其他人没有行边界框，用tesseract库实现这个东西是不可能的，...... ???谢谢

Answer 1

您应该在do循环中移动绘图。

更新

  BitmapBiner = BitmapBiner.copy(Bitmap.Config.RGB_565, true);
  Canvas canvas = new Canvas(BitmapBiner);

  paint.setAlpha(0xA0);
  paint.setColor(Color.RED);
  paint.setStyle(Style.STROKE);
  paint.setStrokeWidth(1);

  do {
    lastUTF8Text = iterator.getUTF8Text(PageIteratorLevel.RIL_SYMBOL);
    lastConfidence = iterator.confidence(PageIteratorLevel.RIL_SYMBOL);
    lastBoundingBox = iterator.getBoundingBox(PageIteratorLevel.RIL_SYMBOL);
    count++;

    // draw bounding box for each character
    Rect r = new Rect(lastBoundingBox[0], lastBoundingBox[1],                       
                      lastBoundingBox[2], lastBoundingBox[3]);
    canvas.drawRect(r, paint);
} while (iterator.next(PageIteratorLevel.RIL_SYMBOL));

对于Android上的TessBaseAPI类型，方法getCharacters（）未定义

1 个答案: