Ejemplo n.º 1
0
  /**
   * Test of extending Tesseract.
   *
   * @throws java.lang.Exception
   */
  @Test
  public void testExtendingTesseract() throws Exception {
    logger.info("Extends Tesseract");
    File imageFile = new File(this.testResourcesDataPath, "eurotext.tif");

    String expResult = "The (quick) [brown] {fox} jumps!\nOver the $43,456.78 <lazy> #90 dog";
    String[] expResults = expResult.split("\\s");

    TessExtension instance1 = new TessExtension();
    instance1.setDatapath(new File(datapath).getPath());
    int pageIteratorLevel = TessPageIteratorLevel.RIL_WORD;
    logger.info(
        "PageIteratorLevel: "
            + Utils.getConstantName(pageIteratorLevel, TessPageIteratorLevel.class));
    List<Word> result = instance1.getTextElements(imageFile, pageIteratorLevel);

    // print the complete result
    for (Word word : result) {
      logger.info(word.toString());
    }

    List<String> text = new ArrayList<String>();
    for (Word word : result.subList(0, expResults.length)) {
      text.add(word.getText());
    }

    assertArrayEquals(expResults, text.toArray());
  }