Ignore images that are contained in others

This commit is contained in:
Dominique Eifländer 2021-05-19 12:54:25 +02:00
parent 8ece616229
commit 0ed7f2b622

View File

@ -102,6 +102,17 @@ public class PdfSegmentationService {
page.setLandscape(isLandscape || isRotated);
page.setPageNumber(pageNumber);
List<PdfImage> mergedList = processImages(stripper.getImages());
List<PdfImage> imagesInImage = new ArrayList<>();
for(PdfImage image: mergedList){
for (PdfImage inner: mergedList){
if(image != inner && image.getPosition().contains(inner.getPosition().getX(), inner.getPosition().getY(), inner.getPosition().getWidth(), inner.getPosition().getHeight())){
imagesInImage.add(inner);
}
}
}
mergedList.removeAll(imagesInImage);
page.setImages(mergedList);
tableExtractionService.extractTables(cleanRulings, page);