net.sourceforge.tess4j.TesseractException Java Examples

protected <SFIRBS> String doRead(SFIRBS from) {
  try {
    String text = "";
    if (from instanceof Mat) {
      Mat img = ((Mat) from).clone();
      if (!img.empty()) {
        img = SXOpenCV.optimize(img, options.factor(), options.resizeInterpolation());
        byte[] bytes = new byte[img.width() * img.height()];
        int n = img.get(0, 0, bytes);
        text = getTesseractAPI().doOCR(img.width(), img.height(), ByteBuffer.wrap(bytes), null, 8);
      } else {
        return "";
      }
    } else {
      BufferedImage bimg = SXOpenCV.optimize(Element.getBufferedImage(from), options.factor(), options.resizeInterpolation());
      text = getTesseractAPI().doOCR(bimg);
    }
    return text.trim().replace("\n\n", "\n");
  } catch (TesseractException e) {
    Debug.error("OCR: read: Tess4J: doOCR: %s", e.getMessage());
    return "";
  }
}
 

public static void main(String[] args) {
    String result = null;
    try {
        File image = new File("src/main/resources/images/baeldung.png");
        Tesseract tesseract = new Tesseract();
        tesseract.setLanguage("spa");
        tesseract.setPageSegMode(1);
        tesseract.setOcrEngineMode(1);
        tesseract.setHocr(true);
        tesseract.setDatapath("src/main/resources/tessdata");
        result = tesseract.doOCR(image, new Rectangle(1200, 200));
    } catch (TesseractException e) {
        e.printStackTrace();
    }
    System.out.println(result);
}
 

@RequestMapping(value = "ocr/v1/convert", method = RequestMethod.POST, consumes = MediaType.APPLICATION_JSON_VALUE, produces = MediaType.APPLICATION_JSON_VALUE)
public Text convertImageToText(@RequestBody final Image image) throws Exception {

    File tmpFile = File.createTempFile("ocr_image", image.getExtension());
    try {
        FileUtils.writeByteArrayToFile(tmpFile, Base64.decodeBase64(image.getImage()));
        Tesseract tesseract = new Tesseract(); // JNA Interface Mapping
        String imageText = tesseract.doOCR(tmpFile);
        LOGGER.debug("OCR Image Text = " + imageText);
        return new Text(imageText);
    } catch (Exception e) {
        LOGGER.error("Exception while converting/uploading image: ", e);
        throw new TesseractException();
    } finally {
        tmpFile.delete();
    }
}
 

@RequestMapping(value = "ocr/v1/upload", method = RequestMethod.POST, consumes = MediaType.APPLICATION_JSON_VALUE, produces = MediaType.APPLICATION_JSON_VALUE)
public Status doOcr(@RequestBody Image image) throws Exception {
    try {
        //FileUtils.writeByteArrayToFile(tmpFile, Base64.decodeBase64(image.getImage()));
        ByteArrayInputStream bis = new ByteArrayInputStream(Base64.decodeBase64(image.getImage()));
        Tesseract tesseract = new Tesseract(); // JNA Interface Mapping
        String imageText = tesseract.doOCR(ImageIO.read(bis));
        image.setText(imageText);
        repository.save(image);
        LOGGER.debug("OCR Result = " + imageText);
    } catch (Exception e) {
        LOGGER.error("TessearctException while converting/uploading image: ", e);
        throw new TesseractException();
    }

    return new Status("success");
}
 

@Override
public String getOCR(File file) {
    Long start = System.currentTimeMillis();
    String result = null;
    try {
        result = instance.doOCR(file);
    } catch (TesseractException e) {
        System.err.println("tessOCR提取图片文字信息失败");
    }
    float time=(System.currentTimeMillis()-start)/(1000f);
    System.out.println("tessOCR提取信息成功，耗时："+time+"s");
    return result;
}
 

private String getText(File file) {
  LOG.info("Peforming OCR on file {}", file);
  try {
    return ocr.doOCR(file).trim();
  } catch (TesseractException exception) {
    LOG.warn("Error performing OCR", exception);
    return null;
  }
}
 

public static void main(String[] args) {
    ITesseract instance = new Tesseract();
    instance.setLanguage("eng");
    try {
        String result;
        result = instance.doOCR(new File("OCRExample.png"));
        System.out.println(result);
    } catch (TesseractException e) {
        System.err.println(e.getMessage());
    }
}
 

/**
 * 识别图片文件中的文字
 * @param instance
 * @param imageFile
 * @return
 */
private static String getOCRText(ITesseract instance, File imageFile){
    String result = null;
    try {
        result = instance.doOCR(imageFile);
    } catch (TesseractException e) {
        e.printStackTrace();
    }
    return result;
}
 

public String read(Picture picture) {
  if (valid) {
    try {
      return tess.doOCR(picture.get());
    } catch (TesseractException e) {
      log.error("read: %s", e.getMessage());
    }
  } else {
    log.error("read: TextFinder not valid");
  }
  return "did not work";
}
 

public void helloWorldTest() throws TesseractException {
  TakesScreenshot driver = mock(TakesScreenshot.class);
  target = new OCR(new Tesseract()).withDriver(driver);
  when(driver.getScreenshotAs(OutputType.FILE)).thenReturn(getPath("helloworld.png"));
  assertThat(target.getText().trim()).isEqualTo("hello world");
}