apache-poi-java-2.java

private void extractImageText(XHTMLContentHandler xhtml, HWPFDocument document) {
  if (Config.inst().getProp(ConfigBool.ENABLE_IMAGE_OCR)) {
    TikaImageHelper helper = new TikaImageHelper(metadata);
    try {
      List<Picture> pictures2 = document.getPicturesTable().getAllPictures();
      for (Picture picture : pictures2) {
        ByteArrayInputStream imageData = new ByteArrayInputStream(
            picture.getContent());
        helper.addImage(ImageIO.read(imageData));
      }
      // TODO: find out page number
      helper.addTextToHandler(xhtml);
    } catch (Exception e) {
      e.printStackTrace();
    } finally {
      if (helper != null) {
        helper.close();
      }
    }
  }
}