Ejemplo n.º 1
0
// Returns an Imagedata containing the image of the given box,
// and ground truth boxes/truth text if available in the input.
// The image is not normalized in any way.
ImageData* Tesseract::GetLineData(const TBOX& line_box,
                                  const GenericVector<TBOX>& boxes,
                                  const GenericVector<STRING>& texts,
                                  int start_box, int end_box,
                                  const BLOCK& block) {
  TBOX revised_box;
  ImageData* image_data = GetRectImage(line_box, block, kImagePadding,
                                       &revised_box);
  if (image_data == NULL) return NULL;
  image_data->set_page_number(applybox_page);
  // Copy the boxes and shift them so they are relative to the image.
  FCOORD block_rotation(block.re_rotation().x(), -block.re_rotation().y());
  ICOORD shift = -revised_box.botleft();
  GenericVector<TBOX> line_boxes;
  GenericVector<STRING> line_texts;
  for (int b = start_box; b < end_box; ++b) {
    TBOX box = boxes[b];
    box.rotate(block_rotation);
    box.move(shift);
    line_boxes.push_back(box);
    line_texts.push_back(texts[b]);
  }
  GenericVector<int> page_numbers;
  page_numbers.init_to_size(line_boxes.size(), applybox_page);
  image_data->AddBoxes(line_boxes, line_texts, page_numbers);
  return image_data;
}
Ejemplo n.º 2
0
// Inserts a list of blobs into the projection.
// Rotation is a multiple of 90 degrees to get from blob coords to
// nontext_map coords, nontext_map_box is the bounds of the nontext_map.
// Blobs are spread horizontally or vertically according to their internal
// flags, but the spreading is truncated by set pixels in the nontext_map
// and also by the horizontal rule line limits on the blobs.
void TextlineProjection::ProjectBlobs(BLOBNBOX_LIST* blobs,
                                      const FCOORD& rotation,
                                      const TBOX& nontext_map_box,
                                      Pix* nontext_map) {
  BLOBNBOX_IT blob_it(blobs);
  for (blob_it.mark_cycle_pt(); !blob_it.cycled_list(); blob_it.forward()) {
    BLOBNBOX* blob = blob_it.data();
    TBOX bbox = blob->bounding_box();
    ICOORD middle((bbox.left() + bbox.right()) / 2,
                  (bbox.bottom() + bbox.top()) / 2);
    bool spreading_horizontally = PadBlobBox(blob, &bbox);
    // Rotate to match the nontext_map.
    bbox.rotate(rotation);
    middle.rotate(rotation);
    if (rotation.x() == 0.0f)
      spreading_horizontally = !spreading_horizontally;
    // Clip to the image before applying the increments.
    bbox &= nontext_map_box;  // This is in-place box intersection.
    // Check for image pixels before spreading.
    TruncateBoxToMissNonText(middle.x(), middle.y(), spreading_horizontally,
                             nontext_map, &bbox);
    if (bbox.area() > 0) {
      IncrementRectangle8Bit(bbox);
    }
  }
}
Ejemplo n.º 3
0
/**
 * Returns the bounding rectangle of the current object at the given level in
 * the coordinates of the working image that is pix_binary().
 * See comment on coordinate system above.
 * Returns false if there is no such object at the current position.
 */
bool PageIterator::BoundingBoxInternal(PageIteratorLevel level,
                                       int* left, int* top,
                                       int* right, int* bottom) const {
  if (Empty(level))
    return false;
  TBOX box;
  PARA *para = NULL;
  switch (level) {
    case RIL_BLOCK:
      box = it_->block()->block->bounding_box();
      break;
    case RIL_PARA:
      para = it_->row()->row->para();
      // explicit fall-through.
    case RIL_TEXTLINE:
      box = it_->row()->row->bounding_box();
      break;
    case RIL_WORD:
      box = it_->word()->word->bounding_box();
      break;
    case RIL_SYMBOL:
      if (cblob_it_ == NULL)
        box = it_->word()->box_word->BlobBox(blob_index_);
      else
        box = cblob_it_->data()->bounding_box();
  }
  if (level == RIL_PARA) {
    PageIterator other = *this;
    other.Begin();
    do {
      if (other.it_->block() &&
          other.it_->block()->block == it_->block()->block &&
          other.it_->row() && other.it_->row()->row &&
          other.it_->row()->row->para() == para) {
        box = box.bounding_union(other.it_->row()->row->bounding_box());
      }
    } while (other.Next(RIL_TEXTLINE));
  }
  if (level != RIL_SYMBOL || cblob_it_ != NULL)
    box.rotate(it_->block()->block->re_rotation());
  // Now we have a box in tesseract coordinates relative to the image rectangle,
  // we have to convert the coords to a top-down system.
  const int pix_height = pixGetHeight(tesseract_->pix_binary());
  const int pix_width = pixGetWidth(tesseract_->pix_binary());
  *left = ClipToRange(static_cast<int>(box.left()), 0, pix_width);
  *top = ClipToRange(pix_height - box.top(), 0, pix_height);
  *right = ClipToRange(static_cast<int>(box.right()), *left, pix_width);
  *bottom = ClipToRange(pix_height - box.bottom(), *top, pix_height);
  return true;
}