Skip to content

Commit

Permalink
put derived images under output fileGrp…
Browse files Browse the repository at this point in the history 
Instead of writing AlternativeImages under a separate
fileGrp (named `OCR-D-IMG-*`), place them under the
output fileGrp (along with PAGE-XML) as well.
Differentiate their file ID further by adding
suffixes (named `.IMG-*`).
  • Loading branch information
@bertsky
bertsky committed Aug 14, 2020
1 parent a7b2d16 commit 6f7f785
Show file tree
Hide file tree
Showing 8 changed files with 99 additions and 149 deletions.
6 changes: 6 additions & 0 deletions CHANGELOG.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,12 @@ Versioned according to [Semantic Versioning](http://semver.org/).

## Unreleased

## [0.1.0] - 2020-08-14

Changed:

* put derived images under output fileGrp, using file ID suffixes

## [0.0.5] - 2020-07-08

Fixed:
Expand Down
2 changes: 1 addition & 1 deletion ocrd_wrap/ocrd-tool.json
View file
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"git_url": "https://github.com/bertsky/ocrd_wrap",
"version": "0.0.5",
"version": "0.1.0",
"tools": {
"ocrd-preprocess-image": {
"executable": "ocrd-preprocess-image",
Expand Down
46 changes: 19 additions & 27 deletions ocrd_wrap/shell.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,9 @@

from ocrd import Processor
from ocrd_utils import (
getLogger, concat_padded,
getLogger,
make_file_id,
assert_file_grp_cardinality,
MIMETYPE_PAGE,
MIME_TO_PIL,
MIME_TO_EXT
Expand All @@ -24,22 +26,13 @@

TOOL = 'ocrd-preprocess-image'
LOG = getLogger('processor.ShellPreprocessor')
FALLBACK_FILEGRP_IMG = 'OCR-D-IMG-PROC'

class ShellPreprocessor(Processor):

def __init__(self, *args, **kwargs):
kwargs['ocrd_tool'] = OCRD_TOOL['tools'][TOOL]
kwargs['version'] = OCRD_TOOL['version']
super(ShellPreprocessor, self).__init__(*args, **kwargs)
if hasattr(self, 'output_file_grp'):
try:
self.page_grp, self.image_grp = self.output_file_grp.split(',')
except ValueError:
self.page_grp = self.output_file_grp
self.image_grp = FALLBACK_FILEGRP_IMG
LOG.info("No output file group for images specified, falling back to '%s'",
FALLBACK_FILEGRP_IMG)

def process(self):
"""Performs coords-preserving image operations via runtime shell calls anywhere.
Expand Down Expand Up @@ -68,9 +61,11 @@ def process(self):
If the shell returns with a failure, skip that segment with an
approriate error message.
Otherwise, add the new image to the workspace with its fileGrp USE
attribute given in the second position of the output fileGrp, or
``OCR-D-IMG-PROC``. Reference it as AlternativeImage in the element,
Otherwise, add the new image to the workspace along with the
output fileGrp, and using a file ID with suffix ``.IMG-``,
and further identification of the input element.
Reference it as AlternativeImage in the element,
adding ``output_feature_added`` to its @comments.
Produce a new PAGE output file by serialising the resulting hierarchy.
Expand All @@ -83,9 +78,11 @@ def process(self):
raise Exception("command parameter requires @INFILE pattern")
if '@OUTFILE' not in command:
raise Exception("command parameter requires @OUTFILE pattern")
assert_file_grp_cardinality(self.input_file_grp, 1)
assert_file_grp_cardinality(self.output_file_grp, 1)

for (n, input_file) in enumerate(self.input_files):
file_id = input_file.ID.replace(self.input_file_grp, self.image_grp)
file_id = make_file_id(input_file, self.output_file_grp)
page_id = input_file.pageId or input_file.ID
LOG.info("INPUT FILE %i / %s", n, page_id)
pcgts = page_from_file(self.workspace.download_file(input_file))
Expand Down Expand Up @@ -158,17 +155,12 @@ def process(self):
"glyph '%s'" % glyph.id, None,
file_id + '_' + glyph.id)

# Use input_file's basename for the new file -
# this way the files retain the same basenames:
file_id = input_file.ID.replace(self.input_file_grp, self.page_grp)
if file_id == input_file.ID:
file_id = concat_padded(self.page_grp, n)
self.workspace.add_file(
ID=file_id,
file_grp=self.page_grp,
file_grp=self.output_file_grp,
pageId=input_file.pageId,
mimetype=MIMETYPE_PAGE,
local_filename=os.path.join(self.page_grp,
local_filename=os.path.join(self.output_file_grp,
file_id + '.xml'),
content=to_xml(pcgts))

Expand All @@ -184,11 +176,11 @@ def _process_segment(self, segment, image, coords, where, page_id, file_id):
in_fd, in_fname = mkstemp(suffix=file_id + MIME_TO_EXT[input_mime])
image.save(in_fname, format=MIME_TO_PIL[input_mime])
# prepare output file name
out_id = file_id + '_' + feature_added
out_fname = os.path.join(self.image_grp,
out_id = file_id + '.IMG-' + feature_added.upper()
out_fname = os.path.join(self.output_file_grp,
out_id + MIME_TO_EXT[output_mime])
if not os.path.exists(self.image_grp):
makedirs(self.image_grp)
if not os.path.exists(self.output_file_grp):
makedirs(self.output_file_grp)
# remove quotation around filename patterns, if any
command = command.replace('"@INFILE"', '@INFILE')
command = command.replace('"@OUTFILE"', '@OUTFILE')
Expand Down Expand Up @@ -226,10 +218,10 @@ def _process_segment(self, segment, image, coords, where, page_id, file_id):
self.workspace.add_file(
ID=out_id,
local_filename=out_fname,
file_grp=self.image_grp,
file_grp=self.output_file_grp,
pageId=page_id,
mimetype=output_mime)
LOG.info("created file ID: %s, file_grp: %s, path: %s",
out_id, self.image_grp, out_fname)
out_id, self.output_file_grp, out_fname)
segment.add_AlternativeImage(AlternativeImageType(
filename=out_fname, comments=features))
48 changes: 18 additions & 30 deletions ocrd_wrap/skimage_binarize.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,9 @@

from ocrd import Processor
from ocrd_utils import (
getLogger, concat_padded,
getLogger,
make_file_id,
assert_file_grp_cardinality,
MIMETYPE_PAGE
)
from ocrd_modelfactory import page_from_file
Expand All @@ -28,25 +30,16 @@

TOOL = 'ocrd-skimage-binarize'
LOG = getLogger('processor.SkimageBinarize')
FALLBACK_FILEGRP_IMG = 'OCR-D-IMG-BIN'

class SkimageBinarize(Processor):

def __init__(self, *args, **kwargs):
kwargs['ocrd_tool'] = OCRD_TOOL['tools'][TOOL]
kwargs['version'] = OCRD_TOOL['version']
super(SkimageBinarize, self).__init__(*args, **kwargs)
if hasattr(self, 'output_file_grp'):
try:
self.page_grp, self.image_grp = self.output_file_grp.split(',')
except ValueError:
self.page_grp = self.output_file_grp
self.image_grp = FALLBACK_FILEGRP_IMG
LOG.info("No output file group for images specified, falling back to '%s'",
FALLBACK_FILEGRP_IMG)

def process(self):
"""Performs binarization of segment or page images with Skimage on the workspace.
"""Performs binarization of segment or page images with scikit-image on the workspace.
Open and deserialize PAGE input files and their respective images,
then iterate over the element hierarchy down to the requested
Expand All @@ -59,19 +52,19 @@ def process(self):
Next, binarize the image according to ``method`` with skimage.
Then write the new image to the workspace with the fileGrp USE given
in the second position of the output fileGrp, or ``OCR-D-IMG-BIN``,
and an ID based on input file and input element.
Then write the new image to the workspace along with the output fileGrp,
and using a file ID with suffix ``.IMG-BIN`` with further identification
of the input element.
Produce a new PAGE output file by serialising the resulting hierarchy.
"""
oplevel = self.parameter['level-of-operation']
assert_file_grp_cardinality(self.input_file_grp, 1)
assert_file_grp_cardinality(self.output_file_grp, 1)

for (n, input_file) in enumerate(self.input_files):
file_id = make_file_id(input_file, self.output_file_grp)
page_id = input_file.pageId or input_file.ID
file_id = input_file.ID.replace(self.input_file_grp, self.image_grp)
if file_id == input_file.ID:
file_id = concat_padded(self.image_grp, n)
LOG.info("INPUT FILE %i / %s", n, page_id)

pcgts = page_from_file(self.workspace.download_file(input_file))
Expand Down Expand Up @@ -114,7 +107,7 @@ def odd(n):
if oplevel == 'page':
self._process_segment(page, page_image, page_coords,
"page '%s'" % page_id, input_file.pageId,
file_id)
file_id + '.IMG-BIN')
continue
regions = page.get_AllRegions(classes=['Text'])
if not regions:
Expand All @@ -125,7 +118,7 @@ def odd(n):
if oplevel == 'region':
self._process_segment(region, region_image, region_coords,
"region '%s'" % region.id, None,
file_id + '_' + region.id)
file_id + '.IMG-BIN_' + region.id)
continue
lines = region.get_TextLine()
if not lines:
Expand All @@ -136,7 +129,7 @@ def odd(n):
if oplevel == 'line':
self._process_segment(line, line_image, line_coords,
"line '%s'" % line.id, None,
file_id + '_' + line.id)
file_id + '.IMG-BIN_' + line.id)
continue
words = line.get_Word()
if not words:
Expand All @@ -147,7 +140,7 @@ def odd(n):
if oplevel == 'word':
self._process_segment(word, word_image, word_coords,
"word '%s'" % word.id, None,
file_id + '_' + word.id)
file_id + '.IMG-BIN_' + word.id)
continue
glyphs = word.get_Glyph()
if not glyphs:
Expand All @@ -157,19 +150,14 @@ def odd(n):
glyph, word_image, word_coords, feature_filter='binarized')
self._process_segment(glyph, glyph_image, glyph_coords,
"glyph '%s'" % glyph.id, None,
file_id + '_' + glyph.id)
file_id + '.IMG-BIN_' + glyph.id)

# Use input_file's basename for the new file -
# this way the files retain the same basenames:
file_id = input_file.ID.replace(self.input_file_grp, self.page_grp)
if file_id == input_file.ID:
file_id = concat_padded(self.page_grp, n)
self.workspace.add_file(
ID=file_id,
file_grp=self.page_grp,
file_grp=self.output_file_grp,
pageId=input_file.pageId,
mimetype=MIMETYPE_PAGE,
local_filename=os.path.join(self.page_grp,
local_filename=os.path.join(self.output_file_grp,
file_id + '.xml'),
content=to_xml(pcgts))

Expand All @@ -196,7 +184,7 @@ def _process_segment(self, segment, image, coords, where, page_id, file_id):
file_path = self.workspace.save_image_file(
image,
file_id,
file_grp=self.image_grp,
file_grp=self.output_file_grp,
page_id=page_id)
segment.add_AlternativeImage(AlternativeImageType(
filename=file_path, comments=features))
Expand Down
48 changes: 18 additions & 30 deletions ocrd_wrap/skimage_denoise.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,9 @@

from ocrd import Processor
from ocrd_utils import (
getLogger, concat_padded,
getLogger,
make_file_id,
assert_file_grp_cardinality,
MIMETYPE_PAGE
)
from ocrd_modelfactory import page_from_file
Expand All @@ -24,25 +26,16 @@

TOOL = 'ocrd-skimage-denoise'
LOG = getLogger('processor.SkimageDenoise')
FALLBACK_FILEGRP_IMG = 'OCR-D-IMG-DEN'

class SkimageDenoise(Processor):

def __init__(self, *args, **kwargs):
kwargs['ocrd_tool'] = OCRD_TOOL['tools'][TOOL]
kwargs['version'] = OCRD_TOOL['version']
super(SkimageDenoise, self).__init__(*args, **kwargs)
if hasattr(self, 'output_file_grp'):
try:
self.page_grp, self.image_grp = self.output_file_grp.split(',')
except ValueError:
self.page_grp = self.output_file_grp
self.image_grp = FALLBACK_FILEGRP_IMG
LOG.info("No output file group for images specified, falling back to '%s'",
FALLBACK_FILEGRP_IMG)

def process(self):
"""Performs binary denoising of segment or page images with Skimage on the workspace.
"""Performs binary denoising of segment or page images with scikit-image on the workspace.
Open and deserialize PAGE input files and their respective images,
then iterate over the element hierarchy down to the requested
Expand All @@ -56,19 +49,19 @@ def process(self):
Next, denoise the image by removing too small connected components
with skimage.
Then write the new image to the workspace with the fileGrp USE given
in the second position of the output fileGrp, or ``OCR-D-IMG-DEN``,
and an ID based on input file and input element.
Then write the new image to the workspace along with the output fileGrp,
and using a file ID with suffix ``.IMG-DEN`` with further identification
of the input element.
Produce a new PAGE output file by serialising the resulting hierarchy.
"""
oplevel = self.parameter['level-of-operation']
assert_file_grp_cardinality(self.input_file_grp, 1)
assert_file_grp_cardinality(self.output_file_grp, 1)

for (n, input_file) in enumerate(self.input_files):
file_id = make_file_id(input_file, self.output_file_grp)
page_id = input_file.pageId or input_file.ID
file_id = input_file.ID.replace(self.input_file_grp, self.image_grp)
if file_id == input_file.ID:
file_id = concat_padded(self.image_grp, n)
LOG.info("INPUT FILE %i / %s", n, page_id)

pcgts = page_from_file(self.workspace.download_file(input_file))
Expand Down Expand Up @@ -106,7 +99,7 @@ def process(self):
if oplevel == 'page':
self._process_segment(page, page_image, page_coords, maxsize,
"page '%s'" % page_id, input_file.pageId,
file_id)
file_id + '.IMG-DEN')
continue
regions = page.get_AllRegions(classes=['Text'])
if not regions:
Expand All @@ -117,7 +110,7 @@ def process(self):
if oplevel == 'region':
self._process_segment(region, region_image, region_coords, maxsize,
"region '%s'" % region.id, None,
file_id + '_' + region.id)
file_id + '.IMG-DEN_' + region.id)
continue
lines = region.get_TextLine()
if not lines:
Expand All @@ -128,7 +121,7 @@ def process(self):
if oplevel == 'line':
self._process_segment(line, line_image, line_coords, maxsize,
"line '%s'" % line.id, None,
file_id + '_' + line.id)
file_id + '.IMG-DEN_' + line.id)
continue
words = line.get_Word()
if not words:
Expand All @@ -139,7 +132,7 @@ def process(self):
if oplevel == 'word':
self._process_segment(word, word_image, word_coords, maxsize,
"word '%s'" % word.id, None,
file_id + '_' + word.id)
file_id + '.IMG-DEN_' + word.id)
continue
glyphs = word.get_Glyph()
if not glyphs:
Expand All @@ -149,19 +142,14 @@ def process(self):
glyph, word_image, word_coords, feature_selector='binarized')
self._process_segment(glyph, glyph_image, glyph_coords, maxsize,
"glyph '%s'" % glyph.id, None,
file_id + '_' + glyph.id)
file_id + '.IMG-DEN_' + glyph.id)

# Use input_file's basename for the new file -
# this way the files retain the same basenames:
file_id = input_file.ID.replace(self.input_file_grp, self.page_grp)
if file_id == input_file.ID:
file_id = concat_padded(self.page_grp, n)
self.workspace.add_file(
ID=file_id,
file_grp=self.page_grp,
file_grp=self.output_file_grp,
pageId=input_file.pageId,
mimetype=MIMETYPE_PAGE,
local_filename=os.path.join(self.page_grp,
local_filename=os.path.join(self.output_file_grp,
file_id + '.xml'),
content=to_xml(pcgts))

Expand All @@ -178,7 +166,7 @@ def _process_segment(self, segment, image, coords, maxsize, where, page_id, file
file_path = self.workspace.save_image_file(
image,
file_id,
file_grp=self.image_grp,
file_grp=self.output_file_grp,
page_id=page_id)
segment.add_AlternativeImage(AlternativeImageType(
filename=file_path, comments=features))
Expand Down
Loading

0 comments on commit 6f7f785

Please sign in to comment.