|
|
|
|
@ -38,9 +38,18 @@ do_linesegmentation() {
|
|
|
|
|
#ocrd workspace validate mets.xml
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
do_ocr() {
|
|
|
|
|
ocrd_tesserocr_recognize_parameters='{ "model": "eng" }' # TODO mods:language + fontident → model
|
|
|
|
|
remove_filegrp OCR-D-OCR-TESS mets.xml
|
|
|
|
|
ocrd-tesserocr-recognize -l DEBUG \
|
|
|
|
|
-m mets.xml -I OCR-D-SEG-LINE -O OCR-D-OCR-TESS \
|
|
|
|
|
-p <(echo $ocrd_tesserocr_recognize_parameters)
|
|
|
|
|
#ocrd workspace validate mets.xml
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
do_fontident
|
|
|
|
|
do_linesegmentation
|
|
|
|
|
do_ocr
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# XXX Multiple calls create multiple identical mets:agent elements
|
|
|
|
|
|
