Marmelator/ocr-service
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 64 Wiki Activity

RED-6126: In the OCRService, OCR Text is not applied to Document

Browse Source 
*removed unnecessary getXObject() call, since it fails for inline_images
This commit is contained in:
Kilian Schuettler 2023-02-13 17:55:02 +01:00
parent 001719a34c
commit 0bc4fea2a5
1 changed files with 1 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
ocr-service-v1/ocr-service-server-v1/src/main/java/com/iqser/red/service/ocr/v1/server/service/ImagePositionRetrievalService.java
View File

@ -15,7 +15,6 @@ import com.pdftron.common.Matrix2D;
import com.pdftron.common.PDFNetException;
import com.pdftron.pdf.Element;
import com.pdftron.pdf.ElementReader;
import com.pdftron.pdf.Image;
import com.pdftron.pdf.PDFDoc;
import com.pdftron.pdf.Page;
import com.pdftron.pdf.Rect;
@ -68,9 +67,8 @@ public class ImagePositionRetrievalService {
while ((element = reader.next()) != null) {
switch (element.getType()) {
case Element.e_image, Element.e_inline_image -> {
Image image = new Image(element.getXObject());
// see everyPointInDashedLineIsImage.pdf TestFile
if (image.getImageHeight() > PIXEL_THRESHOLD || image.getImageWidth() > PIXEL_THRESHOLD) {
if (element.getImageHeight() > PIXEL_THRESHOLD || element.getImageWidth() > PIXEL_THRESHOLD) {
imagePositions.addRect(toRotationAdjustedRect(element.getBBox(), currentPage, mirrorY));
}
}