kba
|
8080bd823c
|
📦 v0.4.0
|
2025-04-07 16:48:57 +02:00 |
|
vahidrezanezhad
|
e2907f67e0
|
'from PIL.Image import Image' causes an error when using Image.new(), and since Image is already imported, this line can be safely commented out.
|
2025-04-06 00:33:36 +02:00 |
|
Robert Sachunsky
|
4339444e47
|
binarization CLI: fix option checks, simplify to asserts, fix dir_in mode
|
2025-04-05 01:21:08 +02:00 |
|
Robert Sachunsky
|
91a340f619
|
CLI: simplify option checks to asserts (also avoid stack trace)
|
2025-04-04 20:42:28 +02:00 |
|
Robert Sachunsky
|
e0a7fde537
|
logger: fix type hint
|
2025-04-04 20:27:15 +02:00 |
|
Robert Sachunsky
|
108ce1f5a1
|
Merge remote-tracking branch 'origin/main' into v3-api-release-foreal
(bad-ass difficult diff diffing)
|
2025-04-04 20:23:23 +02:00 |
|
vahidrezanezhad
|
2e3a29f66b
|
In light mode: To determine whether a main region is a header, I adjusted the ratio to achieve better results.
|
2025-04-04 15:36:31 +02:00 |
|
vahidrezanezhad
|
38a2d60fa2
|
Confidence value for textregions and in the case of not light version is set to zero. This is done to let the pipeline go through. It will be updated to return the correct value in upcomming commits
|
2025-04-03 12:47:27 +02:00 |
|
vahidrezanezhad
|
6b52da227c
|
docorating eynollah with textregion confidence score #135
|
2025-04-03 00:39:21 +02:00 |
|
Robert Sachunsky
|
559d001eef
|
another fix to avoid frequent warnings
|
2025-04-02 05:45:34 +00:00 |
|
Robert Sachunsky
|
dd478279a4
|
CLI: also --overwrite in single-image mode
|
2025-04-02 05:40:21 +00:00 |
|
Robert Sachunsky
|
8159e6336a
|
fix typo (preventing log messages)
|
2025-04-02 00:01:02 +00:00 |
|
Robert Sachunsky
|
2919538382
|
minor fixes to avoid frequent warnings
|
2025-04-01 23:33:26 +00:00 |
|
Robert Sachunsky
|
dcf2ed5e22
|
run: also write out XML in single filename mode
|
2025-04-01 23:13:24 +02:00 |
|
Robert Sachunsky
|
fe77171d45
|
run_single: reduce indentation
|
2025-04-01 22:47:33 +02:00 |
|
Robert Sachunsky
|
79003a083c
|
CLI: ValueError instead of print+exit
|
2025-04-01 22:43:01 +02:00 |
|
Robert Sachunsky
|
e17d34fafa
|
factor run_single() out of run(), simplify kwargs
|
2025-04-01 22:12:24 +02:00 |
|
Robert Sachunsky
|
1a0a1cb00b
|
remove session methods and redundant model loaders
|
2025-04-01 21:15:41 +02:00 |
|
Robert Sachunsky
|
dd51f900b9
|
OCR-D: init Eynollah in 'setup', re-use instance for each page via non-public API
|
2025-04-01 13:02:30 +02:00 |
|
Robert Sachunsky
|
ffeb4a343d
|
Eynollah: remove useless 'pcgts' attr
|
2025-04-01 13:00:41 +02:00 |
|
vahidrezanezhad
|
91b2201b07
|
cnnrnn Ocr: width of input textline image can not be zero!
|
2025-04-01 10:55:40 +02:00 |
|
Robert Sachunsky
|
515b4023f6
|
sbb_binarize: fix missing reference
|
2025-04-01 10:54:36 +02:00 |
|
vahidrezanezhad
|
4de441eaaa
|
OCR prediction is now enabled to integrate results from both RGB and binarized images or to be performed on each individually
|
2025-03-31 21:28:05 +02:00 |
|
vahidrezanezhad
|
b1da0a3327
|
In OCR, the predicted text is now drawn on the image, and the results are saved in a specified directory. This makes it easier to review the predicted output
|
2025-03-31 18:43:14 +02:00 |
|
Robert Sachunsky
|
c01609ff4e
|
allow even more empty imports for optional dependencies
|
2025-03-31 15:57:22 +02:00 |
|
Robert Sachunsky
|
46618f4229
|
allow more empty imports for optional dependencies
|
2025-03-31 14:11:50 +02:00 |
|
Robert Sachunsky
|
4be89910a2
|
CLI: fix arg vs kwarg from merge
|
2025-03-31 02:38:24 +02:00 |
|
Robert Sachunsky
|
9d61acf173
|
simplify
|
2025-03-31 02:02:30 +02:00 |
|
Robert Sachunsky
|
a1068ff2eb
|
OCR-D: move sbb-binarize to ocrd-tool.json, update to v3
|
2025-03-31 01:47:32 +02:00 |
|
Robert Sachunsky
|
c794d4d29f
|
OCR-D: fix typo light_mode→light_version
|
2025-03-31 01:46:29 +02:00 |
|
Robert Sachunsky
|
4338259ca1
|
OCR-D: ensure page image gets replaced in result as well if not the original file
|
2025-03-31 01:17:14 +02:00 |
|
Robert Sachunsky
|
55969b0173
|
OCR-D: add docstring
|
2025-03-31 01:15:26 +02:00 |
|
Robert Sachunsky
|
6d02e90570
|
OCR-D: restrict max_workers=1
|
2025-03-31 01:14:54 +02:00 |
|
Robert Sachunsky
|
efd3fa6775
|
allow empty imports for optional dependencies
|
2025-03-31 00:32:26 +02:00 |
|
Robert Sachunsky
|
238132e260
|
use 'image_filename' for pseudo-iteration outside 'dir_in' mode
|
2025-03-31 00:31:49 +02:00 |
|
Robert Sachunsky
|
af4e2a4ffc
|
do not require 'dir_out' outside 'dir_in' mode
|
2025-03-31 00:31:09 +02:00 |
|
Robert Sachunsky
|
ea136e3ddd
|
'overwrite' check: only in 'dir_in' mode
|
2025-03-31 00:30:06 +02:00 |
|
Robert Sachunsky
|
1f4a17b60d
|
Merge remote-tracking branch 'origin/machine_based_reading_order_integration' into v3-api
|
2025-03-30 21:21:59 +02:00 |
|
Robert Sachunsky
|
edf924c2cb
|
ocrd-tool: add dockerhub
|
2025-03-30 19:47:25 +02:00 |
|
vahidrezanezhad
|
9b04688ebc
|
The rotate_image function has been updated. Additionally, the reading order is now correct in the case of the light version, provided that slope_deskew exceeds the slope_threshold.
|
2025-03-30 15:34:27 +02:00 |
|
vahidrezanezhad
|
cf40f9ecc5
|
The rotate_image function produces the exact same rotation as Imutils. Therefore, there is no need to retain the remove-imutils-1 branch.
|
2025-03-28 20:58:32 +01:00 |
|
vahidrezanezhad
|
f756b08c9b
|
Revert "replace usages of imutils with opencv equivalents"
|
2025-03-28 14:57:40 +01:00 |
|
vahidrezanezhad
|
52c605185a
|
Merge pull request #146 from qurator-spk/remove-imutils-1
replace usages of `imutils` with opencv equivalents
|
2025-03-28 11:10:25 +01:00 |
|
vahidrezanezhad
|
6f36c7177f
|
For OCR, the splitting ratio of text lines is adjusted
|
2025-03-27 18:24:47 +01:00 |
|
cneud
|
181c0c584f
|
bbox rotation with opencv
|
2025-03-26 22:25:22 +01:00 |
|
cneud
|
eaff9e3537
|
Merge branch 'main' into remove-imutils-1
|
2025-03-26 20:16:46 +01:00 |
|
vahidrezanezhad
|
7df0427b04
|
In the context of OCR, if Page-XML files already contain text, the new predicted text will replace the existing text.
|
2025-03-26 18:42:06 +01:00 |
|
vahidrezanezhad
|
370d44a66b
|
Slope deskew in the light version is set to zero because when the slope_deskew value exceeds the slope_threshold, the reading order becomes incorrect. This issue needs to be addressed. Additionally, the textlines order within text region in the light version was reversed, and this has been corrected.
|
2025-03-26 10:45:34 +01:00 |
|
vahidrezanezhad
|
d3a4c06e7f
|
This commit enables the export of cropped text line images along with their corresponding texts from a Page-XML file. These exported text line images and texts can be utilized for training a text line-based OCR model.
|
2025-03-20 18:21:44 +01:00 |
|
vahidrezanezhad
|
c8b8529951
|
For the CNN-RNN OCR model, long text lines are split into two segments
|
2025-03-17 19:50:58 +01:00 |
|