Commit graph

488 commits

Author SHA1 Message Date
vahidrezanezhad
70772d4104 binarization as a standalone command 2024-10-21 23:46:38 +02:00
vahidrezanezhad
f93fa12441 doing more multiprocessing in order to make the process faster 2024-10-18 09:14:42 +02:00
vahidrezanezhad
3ef4eac24c textlines of textregions are extracted in a faster way + early layout for all documents is done with no patches model and on rgb input 2024-10-17 19:12:28 +02:00
vahidrezanezhad
1da4b7f589 updating light version 2024-10-07 10:55:10 +02:00
vahidrezanezhad
543ed4bc38 -light version need -tll to be enabled otherwise the process will be ended. 2024-10-02 14:09:13 +02:00
vahidrezanezhad
ab63d5ba40 updating light version features 2024-09-30 21:28:39 +02:00
vahidrezanezhad
1774076f4a updating light version. Remove textlines or textregion contours inside a bigger one 2024-09-30 16:10:29 +02:00
vahidrezanezhad
ad32316217 updating light version 2024-09-27 20:59:01 +02:00
vahidrezanezhad
133091137d dilation of textregions and marginals are accomplished 2024-09-27 13:57:01 +02:00
vahidrezanezhad
95effe54a0 updating textregions dilation 2024-09-25 20:00:53 +02:00
vahidrezanezhad
b33739adee parametriyation in the case of textline contours dilation is accomplished 2024-09-24 16:06:27 +02:00
vahidrezanezhad
6626dc6866 updating textline dilation parameters 2024-09-23 15:50:37 +02:00
vahidrezanezhad
62f8ae4860 updating dilation of textlines and text regions 2024-09-23 14:03:07 +02:00
vahidrezanezhad
7f08458436 dilation of text regions without opencv 2024-09-21 14:39:54 +02:00
vahidrezanezhad
5d680136a4 updating light version 2024-09-21 01:04:28 +02:00
vahidrezanezhad
b9e8959c4a update of light versions 2024-09-20 16:33:13 +02:00
vahidrezanezhad
2d18739d9b postprocessing of textline contour dilation + skip layout and reading order passed as an argument 2024-09-20 15:08:09 +02:00
vahidrezanezhad
5a07cd9cfa the most effective version of contours dilation without opencv and all at once 2024-09-19 16:21:55 +02:00
vahidrezanezhad
a1f1f98de3 updating scaling contours 2024-09-18 00:08:54 +02:00
vahidrezanezhad
21380fc870 scaling contours without dilation 2024-09-17 15:06:41 +02:00
vahidrezanezhad
1b18ae874b passing number of columns as an argument 2024-09-13 00:52:06 +02:00
vahidrezanezhad
2c93904985 avoiding double binarization 2024-09-12 17:35:28 +02:00
vahidrezanezhad
f0b49073b7 adding option for textline detection in printspace 2024-09-03 23:10:38 +02:00
vahidrezanezhad
c3a4a1bba7 resolving issue #110 in a better way 2024-09-03 13:14:10 +02:00
vahidrezanezhad
0f87974b0c writing drop capitals in xml output + and may resolve issue #110 2024-09-02 16:21:07 +02:00
vahidrezanezhad
93005959e5 inference batch size debugged 2024-08-27 18:13:46 +02:00
vahidrezanezhad
7ae6a8776f ignoring dpi check by light version 2024-08-26 16:02:10 +02:00
vahidrezanezhad
04e79002b3 making light version faster for 1 and 2 columns images 2024-08-24 12:54:19 +02:00
vahidrezanezhad
c10a525675 inference with batch size bigger than 1 2024-08-23 02:18:16 +02:00
cneud
4f8210de71 update Makefile model location 2024-08-15 23:23:48 +02:00
vahidrezanezhad
6f4205ba49 update pyproject.toml 2024-08-15 16:08:45 +02:00
vahidrezanezhad
74eac4dacc dtype = object in the case of length 1 arise error 2024-08-15 13:50:36 +02:00
vahidrezanezhad
4c50479cb8 pyproject.toml may work for ocrd 2024-08-14 15:28:36 +02:00
vahidrezanezhad
53fd5fb2a5 resolving #106 for pyproject.toml test 2024-08-14 14:42:37 +02:00
vahidrezanezhad
e976778796 testing pyproject.toml 2024-08-14 14:33:01 +02:00
vahidrezanezhad
00bf2b64d0 1&2 column images only printspace 2024-08-07 19:07:54 +02:00
vahidrezanezhad
be144db9f8 updating 1&2 columns images + full layout 2024-08-07 18:13:10 +02:00
vahidrezanezhad
a62ae370c3 new full layout model and early layout for 1&2 column images are integrated - light version 2024-08-07 02:21:01 +02:00
vahidrezanezhad
5144668834 ocr engine first integration 2024-07-17 10:01:37 +02:00
vahidrezanezhad
eac18c553d machine based reading order as an argument 2023-12-13 01:44:51 +01:00
vahidrezanezhad
941d87328a machine based reading order & works for not full layout case 2023-10-20 11:19:30 +02:00
vahidrezanezhad
59c0d90e5a machine based reading order inference & optimized algorithm 2023-10-20 10:17:46 +02:00
vahidrezanezhad
49c93149a4 machine based reading order inference with a variable batch size 2023-10-20 10:01:28 +02:00
vahidrezanezhad
5fdc6d4fa4 integration of machine based reading order detection 2023-10-14 09:05:05 +02:00
vahidrezanezhad
fc9e9cc29f
Merge pull request #117 from qurator-spk/tf-2.12-or-greater
Update tensorflow
2023-09-29 05:30:19 -04:00
cneud
4254ce3bdb Update README.md 2023-09-26 18:54:14 +02:00
cneud
56934c876a remove duplicate test for Python 3.8 2023-09-26 18:53:10 +02:00
Clemens Neudecker
6c65fc4dfe
Update config.yml 2023-09-26 18:33:05 +02:00
Clemens Neudecker
9d3a1a5b76
Update test-eynollah.yml 2023-09-26 18:32:20 +02:00
Clemens Neudecker
03bfd7a390
Update requirements.txt
Update to `tensorflow>=2.12` (drops Python 3.7 support)
* fix #114 
* fix #115

Tested by @vahidrezanezhad @cneud
2023-09-26 18:16:20 +02:00