Commit graph

57 commits

Author SHA1 Message Date
b-vr103
1446d7c662 getting robust and doing sth for verticals 2019-12-13 18:04:04 +01:00
b-vr103
3941f2f17d gettin robust and doing sth for verticals 2019-12-13 17:49:19 +01:00
f90b3cfa86 🔊 sbb_textline_detector: In OCR-D interface, warn if overwriting existing segmentation 2019-12-11 13:54:29 +01:00
11c0e9cee5 🐛 sbb_textline_detector: Do not print PAGE output to stdout
ocrd-sbb-textline-detector uses ocrd_page's parse() to parse XML input,
which writes the XML to stdout by default.

Fix this by silencing it using parse()'s silence=True.
2019-12-11 12:39:50 +01:00
wrznr
4fc57d7756 Assign page id 2019-12-10 16:59:45 +01:00
wrznr
9e9163e852 Simplify the iteration over files in the input file group 2019-12-10 16:55:43 +01:00
6e0decb5ec
Merge pull request #12 from kba/rename-tool
Rename ocrd_sbb.. to ocrd-sbb... in ocrd_cli.py, ht @bertsky
2019-12-09 16:50:27 +01:00
5fb30a7a1f Revert "Merge branch 'master' of https://github.com/qurator-spk/sbb_textline_detector"
This reverts commit 417b9235d5, reversing
changes made to a74974b7b6.
2019-12-09 15:11:25 +01:00
Konstantin Baierer
cf6381c148 Rename ocrd_sbb.. to ocrd-sbb... in ocrd_cli.py, ht @bertsky 2019-12-09 13:02:44 +01:00
Clemens Neudecker
51e241fd84
Merge pull request #5 from cneud/cneud-fix-typos
Fix typos
2019-12-06 19:45:14 +01:00
Clemens Neudecker
12c07f389d
Merge pull request #7 from cneud/cneud-fix-docstring
fix docstring
2019-12-06 19:44:34 +01:00
Clemens Neudecker
29870f26e1
Merge pull request #4 from cneud/cneud-PAGE2019
PAGE2019
2019-12-06 19:44:04 +01:00
Konstantin Baierer
b6ca1a7c53 kebab-case snake_case executable, fix #9 2019-12-06 18:26:09 +01:00
Clemens Neudecker
6c0bfba686
fix typos 2019-12-06 02:21:04 +01:00
Clemens Neudecker
c8bc468628
fix docstring 2019-12-06 00:40:05 +01:00
Clemens Neudecker
e696a068cb
Fix typos 2019-12-06 00:20:34 +01:00
Clemens Neudecker
d90dad48fd
PAGE2019 2019-12-05 22:24:28 +01:00
Rezanezhad, Vahid
19116091f9 Update config_params.json 2019-12-05 14:05:55 +01:00
af5cbe9052 🐛 sbb_textline_detector: Fix making the output file id 2019-12-04 11:42:45 +01:00
Rezanezhad, Vahid
2112bb18c6 fixed the bug: local variable 't4' referenced before assignment 2019-11-29 11:29:12 +01:00
Rezanezhad, Vahid
a11f6740cb Update main.py - robust deskewing and better page extraction 2019-11-28 16:19:44 +01:00
Rezanezhad, Vahid
0182b7087f remove multiprocessing bug 2019-11-20 14:05:15 +01:00
8fa7179560 🐛 sbb_textline_detector: Disable multiprocessing to fix race condition
Lines were sorted in the wrong regions. Work around this by disabling
multiprocessing until a proper fix is done.
2019-11-20 09:50:29 +01:00
4aed06a325 sbb_textline_detection: Preserve input PAGE info by merging segmentation results
ocrd_sbb_textline_detection used the output XML by main.py as is, and
– by doing this – threw away any input data from the input PAGE,
including the critical pc:AlternativeImage and the less important
pc:MetadataItem.

Fix this by merging the segmentation results into a file created from
the input file.

Also add a pc:MetadataItem processingStep about the segmentation
operation.
2019-11-19 15:08:53 +01:00
4fb3e70ef6 🧹 sbb_textline_detector: Do not create empty/space-only TextEquivs (again) 2019-11-19 11:08:41 +01:00
bf41a29e7b 🐛 sbb_textline_detector: Do not hardcode Created/LastChange elements 2019-11-19 11:05:18 +01:00
fbd21cdb81 🧹 sbb_textline_detector: Do not create empty/space-only TextEquivs (again) 2019-11-19 10:59:41 +01:00
Rezanezhad, Vahid
2d6dd92b31 Update main.py 2019-11-04 11:10:17 +01:00
Rezanezhad, Vahid
9f97f34255 Update main.py 2019-10-31 17:36:21 +01:00
Rezanezhad, Vahid
8c954a6c7a Update main.py 2019-10-31 17:08:35 +01:00
Rezanezhad, Vahid
6714481556 Update main.py 2019-10-31 10:54:57 +01:00
Rezanezhad, Vahid
719824f19d Update main.py 2019-10-30 13:37:54 +01:00
f94511a1d8 Merge branch 'master' of code.dev.sbb.berlin:qurator/mono-repo 2019-10-25 18:11:17 +02:00
4f28cd905a 🧹 sbb_textline_detector: Do not create empty/space-only TextEquivs
ocrd_tesserocr or ocrd_cis complain about already existing text if
empty/space-only TextEquivs elements exist after segmentation. Also, it
does not make sense to create them in a segmentation step.

Fix by removing the code generating the elements.
2019-10-25 18:08:31 +02:00
Rezanezhad, Vahid
00929ab391 Update main.py 2019-10-25 14:39:37 +02:00
f0dd955606 Merge branch 'master' of code.dev.sbb.berlin:qurator/mono-repo 2019-10-25 14:20:44 +02:00
2528573b4f sbb_textline_detector: Allow PAGE input in OCR-D interface
Previous OCR-D processors may output PAGE files instead of image files.
Resolve images file from PAGE files if necessary.
2019-10-25 14:16:09 +02:00
Rezanezhad, Vahid
d8e04e3de4 memory leakage is removed. New deskewing methid is integrated. 2019-10-25 14:07:36 +02:00
Rezanezhad, Vahid
47d972b459 Update main.py 2019-10-22 13:27:23 +02:00
103cfa0565 Merge branch 'master' of code.dev.sbb.berlin:qurator/mono-repo 2019-10-18 13:18:27 +02:00
7884ab93c6 🧹 sbb_textline_detector: Destroy Keras session at the end of a run() to free up memory 2019-10-18 10:59:41 +02:00
5d440857e7 🧹 sbb_textline_detector: Delete textline session/model after using it 2019-10-18 10:59:01 +02:00
cneud
4201fa7d0f sbb_textline_detector: typo (polugons --> polygons) 2019-10-16 18:52:35 +02:00
9b2c415125 🐛 sbb_textline_detector: Use the correct image filename in the output PAGE 2019-10-15 18:08:47 +02:00
Rezanezhad, Vahid
1702472401 Update main.py 2019-10-15 14:32:40 +02:00
Rezanezhad, Vahid
ca9f47eb20 Update main.py 2019-10-15 14:03:09 +02:00
Rezanezhad, Vahid
419beed836 Update main.py 2019-10-15 13:24:27 +02:00
2199bf0d8c 🧹 sbb_textline_detector: Remove extra .xml suffix from METS file id 2019-10-11 14:37:51 +02:00
b4bef6460c 🐛 sbb_textline_detector: Use the correct image filename in the output PAGE 2019-10-11 13:15:33 +02:00
1c7d45d3d0 ♻ sbb_textline_detector: Remove redundant and wrongly named parameter dir_of_image 2019-10-11 13:14:57 +02:00