qurator-spk/ocrd-galley
1
0
Fork 0
mirror of https://github.com/qurator-spk/ocrd-galley.git synced 2025-10-31 03:04:12 +01:00
Code Issues Projects Releases Wiki Activity

✨ Use ocrd_olena for binarization

Browse source
This commit is contained in:
Gerber, Mike 2019-10-21 17:04:06 +02:00
parent 63c364207c
commit 7025d960b4
2 changed files with 16 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

7
my_ocrd_workflow
View file

@ -36,8 +36,9 @@ do_binarization() {
# Binarize the images
remove_filegrp OCR-D-IMG-BIN mets.xml
ocrd-kraken-binarize -l $LOG_LEVEL \
-m mets.xml -I OCR-D-IMG -O OCR-D-IMG-BIN
ocrd-olena-binarize -l $LOG_LEVEL \
-m mets.xml -I OCR-D-IMG -O OCR-D-IMG-BIN \
-p <(echo '{"impl": "sauvola-ms-split"}')
}
do_fontident() {
@ -143,7 +144,7 @@ page_fix_image_references_to_bin() {
for file in `ocrd workspace find -G $filegrp`; do
# Arrays with filenames to the images
imgs=(`ocrd workspace find -G OCR-D-IMG`)
imgs_bin=(`ocrd workspace find -G OCR-D-IMG-BIN`)
imgs_bin=(`ocrd workspace find -G OCR-D-IMG-BIN -m image/png`)
# Change all image references to point to the corresponding binarized image
for i in ${!imgs[@]}; do