qurator-spk/sbb_textline_detection
1
0
Fork 0
mirror of https://github.com/qurator-spk/sbb_textline_detection.git synced 2025-06-09 03:40:18 +02:00
Code Issues Releases Wiki Activity

✨ sbb_textline_detector: Allow PAGE input in OCR-D interface

Browse source 
Previous OCR-D processors may output PAGE files instead of image files.
Resolve images file from PAGE files if necessary.
This commit is contained in:
Gerber, Mike 2019-10-25 14:16:09 +02:00
parent 47d972b459
commit 2528573b4f
1 changed files with 17 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

22
qurator/sbb_textline_detector/ocrd_cli.py
View file

@ -4,7 +4,9 @@ import os
import click import click
from ocrd import Processor from ocrd import Processor
from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor from ocrd.decorators import ocrd_cli_options, ocrd_cli_wrap_processor
from ocrd_utils import concat_padded, getLogger from ocrd_modelfactory import page_from_file
from ocrd_models import OcrdFile
from ocrd_utils import concat_padded, getLogger, MIMETYPE_PAGE
from pkg_resources import resource_string from pkg_resources import resource_string
from qurator.sbb_textline_detector import textlineerkenner from qurator.sbb_textline_detector import textlineerkenner
@ -32,12 +34,22 @@ class OcrdSbbTextlineDetectorRecognize(Processor):
file_id = concat_padded(self.output_file_grp, n) file_id = concat_padded(self.output_file_grp, n)
return file_id return file_id
def _resolve_image_file(self, input_file: OcrdFile) -> str:
if input_file.mimetype == MIMETYPE_PAGE:
pcgts = page_from_file(self.workspace.download_file(input_file))
page = pcgts.get_Page()
image_file = page.imageFilename
else:
image_file = input_file.local_filename
return image_file
def process(self): def process(self):
for n, page_id in enumerate(self.workspace.mets.physical_pages): for n, page_id in enumerate(self.workspace.mets.physical_pages):
image_file = self.workspace.mets.find_files(fileGrp=self.input_file_grp, pageId=page_id)[0] input_file = self.workspace.mets.find_files(fileGrp=self.input_file_grp, pageId=page_id)[0]
log.info("INPUT FILE %i / %s", n, image_file) log.info("INPUT FILE %i / %s", n, input_file)
file_id = self._make_file_id(image_file, self.output_file_grp, n) file_id = self._make_file_id(input_file, self.output_file_grp, n)
image_file = self._resolve_image_file(input_file)
# Process the files # Process the files
try: try:
@ -46,7 +58,7 @@ class OcrdSbbTextlineDetectorRecognize(Processor):
pass pass
model = self.parameter['model'] model = self.parameter['model']
x = textlineerkenner(image_file.local_filename, self.output_file_grp, file_id, model) x = textlineerkenner(image_file, self.output_file_grp, file_id, model)
x.run() x.run()
self.workspace.add_file( self.workspace.add_file(