Commit graph

  • 75555b8e0c
    Merge 42a3cc2335 into a2359ea4c4 Robert Sachunsky 2025-08-28 15:25:12 +02:00
  • 87938fe42b
    Merge fdcae8dd6e into a2359ea4c4 vahidrezanezhad 2025-08-28 11:37:36 +02:00
  • fdcae8dd6e eynollah ocr: support using either a specific model name or a models directory (default model) vahidrezanezhad 2025-08-28 11:30:59 +02:00
  • 7dd281267d Marginals are divided into left and right, and written from top to bottom. vahidrezanezhad 2025-08-26 22:38:03 +02:00
  • 03e2a7dfc1
    Merge fd6a6495a2 into a2359ea4c4 Robert Sachunsky 2025-08-21 13:01:31 +02:00
  • fd6a6495a2 increase dilatation: textregions/lines (5→6), seplines (0→1) Robert Sachunsky 2025-08-21 13:00:31 +02:00
  • 8be52fb143 refactor shapely converisons into contour2polygon / polygon2contour, also handle heterogeneous geometries Robert Sachunsky 2025-08-21 12:59:03 +02:00
  • 8b5f90e243 move dilate_*_contours to .utils.contour, rename dilate_textregions_contours_textline_version → dilate_textline_contours Robert Sachunsky 2025-08-21 01:42:46 +02:00
  • 244772f086 filter_contours_area_of_image*: also ensure validity here Robert Sachunsky 2025-08-21 01:33:16 +02:00
  • 42474afa4b rename *lines_xml → *seplines for clarity Robert Sachunsky 2025-08-21 01:32:32 +02:00
  • b610fe07a6 check_any_text_region_in_model_one_is_main_or_header_light: return original instead of resampled contours Robert Sachunsky 2025-08-21 01:05:15 +02:00
  • 3d53070b90 avoid creating invalid polygons via rounding Robert Sachunsky 2025-08-21 01:03:46 +02:00
  • 277d00579e get_textregion_contours_in_org_image_light: no back rotation, drop slope_first (always 0) Robert Sachunsky 2025-08-20 14:28:14 +02:00
  • b6d1c43a85 dilate_textregions_contours_textline_version: simplify (via shapely's Polygon.buffer()), ensure validity Robert Sachunsky 2025-08-20 14:26:14 +02:00
  • 6c442c9ae9 separate_lines/do_work_of_slopes: skip if crop is empty Robert Sachunsky 2025-08-19 22:56:36 +02:00
  • e9a6ff5d81 return_boxes_of_images_by_order_of_reading_new: simplify, avoid changing dtype during np.append Robert Sachunsky 2025-08-19 20:09:09 +02:00
  • f994ea5f0b dilate_textregions_contours: simplify (via shapely's Polygon.buffer()), ensure validity Robert Sachunsky 2025-08-19 11:58:45 +02:00
  • 342b87bf33
    Merge 8ebba5ac04 into a2359ea4c4 michalbubula 2025-08-18 10:16:10 +02:00
  • 8dc2fab9fa reading order on given layout vahidrezanezhad 2025-08-18 02:31:13 +02:00
  • b83dcc729b
    Merge 2996fc8b30 into a2359ea4c4 vahidrezanezhad 2025-08-15 18:41:39 +02:00
  • a2359ea4c4
    Merge pull request #171 from bertsky/ocrd-machine-based-ro main Clemens Neudecker 2025-08-15 18:40:13 +02:00
  • 21615a986d OCR-D processor: expose reading_order_machine_based Robert Sachunsky 2025-08-13 14:14:37 +02:00
  • 8ebba5ac04 add feedback to command line interface michalbubula 2025-08-12 16:21:15 +02:00
  • 20614d1678 avoiding float in range vahidrezanezhad 2025-08-12 12:50:15 +02:00
  • 5db3e9fa64 deskewing with faster multiprocessing vahidrezanezhad 2025-08-08 11:32:02 +02:00
  • a0c19c57be use the latest ocr model with balanced fraktur-antiqua training dataset vahidrezanezhad 2025-08-05 14:22:22 +02:00
  • 0803881f36 threshold for textline ocr + new ocr model vahidrezanezhad 2025-07-25 13:18:38 +02:00
  • 2996fc8b30
    Merge pull request #166 from qurator-spk/updating_readme_for_eynollah_use_cases-cli updating_readme_for_eynollah_use_cases Clemens Neudecker 2025-07-24 15:30:57 +02:00
  • fd0595f920
    Update Makefile updating_readme_for_eynollah_use_cases-cli vahidrezanezhad 2025-07-24 13:52:38 +02:00
  • 611a521045
    Merge 3b1886973d into 6b8893b188 Konstantin Baierer 2025-07-24 14:19:10 +10:00
  • da141bb42e resolving tests error vahidrezanezhad 2025-07-23 16:44:17 +02:00
  • 6b8893b188
    Merge pull request #167 from qurator-spk/ocrd-fixes vahidrezanezhad 2025-07-22 14:46:25 +02:00
  • d968a306e4 should merged text for the whole page be written in xml? vahidrezanezhad 2025-07-21 14:50:05 +02:00
  • 920705c3b1 update model names vahidrezanezhad 2025-07-21 10:54:20 +02:00
  • e0f4a007e4 ocr model renamed - image text font for ocr result is now using Charis-7.000 font (downloaded from here https://software.sil.org/charis/download/) vahidrezanezhad 2025-07-16 14:00:12 +02:00
  • e54ebaa23e ocr: make sure that image height or width is not zero vahidrezanezhad 2025-07-03 15:24:52 +02:00
  • 59ea493803 decorated with confidence value for cnnrnn ocr model vahidrezanezhad 2025-07-03 11:50:47 +02:00
  • 42a3cc2335 cv2pil: limit color depth on output imgs Robert Sachunsky 2025-06-13 01:21:25 +02:00
  • dadb879376 pil2cv: allow (and drop) alpha channels on input imgs Robert Sachunsky 2025-06-12 20:55:37 +02:00
  • b7b218ff11 OCR-D processor: same behavior as standalone wrt light_version/textline_light ocrd-fixes kba 2025-06-12 15:30:17 +02:00
  • c194a20c9c Fixed duplicate textline_light assignments (true and false) in the OCR-D framework for the Eynollah light version, which caused rectangles to be used instead of contours for textlines vahidrezanezhad 2025-06-11 18:57:08 +02:00
  • 32889ef1e0 adapt binarization CLI according to #156 kba 2025-06-12 13:57:41 +02:00
  • 9b4e78c55c
    Fixed duplicate textline_light assignments (true and false) in the OCR-D framework for the Eynollah light version, which caused rectangles to be used instead of contours for textlines vahidrezanezhad 2025-06-11 18:57:08 +02:00
  • 065f1f9a93 Fix: Resolved OCR bug when text region type is undefined vahidrezanezhad 2025-06-02 18:21:33 +02:00
  • 7996afac69 image enhancer updated vahidrezanezhad 2025-06-01 22:44:50 +02:00
  • d14bd162ca saving enhanced image in org or scaled resolution vahidrezanezhad 2025-06-01 22:10:13 +02:00
  • cc36694dfd image enhancer is integrated vahidrezanezhad 2025-06-01 15:53:04 +02:00
  • 928a548b70 Parametrize OCR for handling curved lines vahidrezanezhad 2025-05-31 01:09:14 +02:00
  • 48285ce3f5 updating ocr vahidrezanezhad 2025-05-28 01:17:21 +02:00
  • b93fc112bf updating ocr vahidrezanezhad 2025-05-27 23:45:22 +02:00
  • 0f154c605a strings alignment function is added + new changes needed for prediction with both bin and rgb inputs is implemented vahidrezanezhad 2025-05-25 21:44:36 +02:00
  • 097520bfd2 rnn ocr for all layout textregion types vahidrezanezhad 2025-05-25 03:33:54 +02:00
  • 27c4b0d0e0 Drop capitals are written separately and are not attached to their corresponding text line. The OCR use case also supports single-image input. vahidrezanezhad 2025-05-25 01:12:58 +02:00
  • adcf03c7b7 enhancing ocr vahidrezanezhad 2025-05-23 18:06:53 +02:00
  • d4f6e10251 commit 21ec4fb is picked + rnn ocr at the same time with segmentation + enhancement of mb reading order vahidrezanezhad 2025-05-23 15:55:03 +02:00
  • a0647eff93 enhancing curved lines OCR vahidrezanezhad 2025-05-21 17:42:44 +02:00
  • f94fc9973b Implement hyphenated textline merging in OCR engine and a bug fixed for curved textline OCR vahidrezanezhad 2025-05-21 14:39:31 +02:00
  • c0835665a9 ocr for curved lines vahidrezanezhad 2025-05-20 19:01:52 +02:00
  • 848156dd9d mb reading order now can be done faster. Text regions are clustered using dilation, and mb reading order needs to be implemented for fewer regions vahidrezanezhad 2025-05-20 16:51:08 +02:00
  • 7a34bbb493 enhancing marginal detection for light version vahidrezanezhad 2025-05-18 02:48:05 +02:00
  • 0819730355 marginals detection enhanced for light version vahidrezanezhad 2025-05-15 15:33:50 +02:00
  • adee1dc55c enhancement for vertical textlines vahidrezanezhad 2025-05-15 00:45:22 +02:00
  • 7a22e51f5d resolve some comments from review cneud 2025-05-14 21:56:03 +02:00
  • a9cdd56e9a enhance ocr for vertical textlines vahidrezanezhad 2025-05-14 18:34:58 +02:00
  • 1ccd3fb7cf Accurately writing text line contours into xml file when the deskewing exceeds 45 degrees and the text line is in light mode vahidrezanezhad 2025-05-13 15:53:05 +02:00
  • 07f5b52fa7 The initial attempt at reading heavily deskewed or vertically aligned lines. vahidrezanezhad 2025-05-13 14:40:57 +02:00
  • 02a679a145 I have tried to address the issues #163 and #161 . The changes have also improved marginal detection and enhanced the isolation of headers. vahidrezanezhad 2025-05-12 00:10:18 +02:00
  • 21ec4fbfb5 The text region coordinates are now correctly written into the XML output when using the skip layout and reading order option vahidrezanezhad 2025-05-07 14:04:01 +02:00
  • 83211ae684 In the case of skip_layout_and_reading_order, the confidence value was not set correctly, leading to an error while writing to the XML file. vahidrezanezhad 2025-05-07 12:33:03 +02:00
  • 3dcbb20cac
    Merge pull request #159 from bertsky/main Clemens Neudecker 2025-05-06 15:14:06 +02:00
  • 5d447abcc4 let to add dataset abbrevation to extracted textline images and text vahidrezanezhad 2025-05-03 02:59:16 +02:00
  • 8c8fa461bb machine based model name changed to public one vahidrezanezhad 2025-05-02 12:57:26 +02:00
  • a4defbb04d inference batch size for ocr is passed as an argument vahidrezanezhad 2025-05-02 12:53:33 +02:00
  • fd375e15d5 adding space between splitted textline predicted text in the case of trocr vahidrezanezhad 2025-05-02 01:02:32 +02:00
  • 5c8084a397 displaying detexted text on an image is provided for trocr case vahidrezanezhad 2025-05-02 00:30:36 +02:00
  • e9179e1d34 docker: use latest core base stage Robert Sachunsky 2025-05-02 00:13:06 +02:00
  • f8b4d29a59 docker: prepackage ocrd-all-module-dir.json Robert Sachunsky 2025-05-02 00:13:11 +02:00
  • e2da7a6239 Fix model name to return the correct machine-based model name vahidrezanezhad 2025-04-30 16:06:29 +02:00
  • b227736094 Fix OCR text cleaning to correctly handle 'U', 'K', and 'N' starting sentence; update text line splitting size vahidrezanezhad 2025-04-30 16:04:34 +02:00
  • 4cb4414740 Resolve remaining issue with #158 and resolving #124 vahidrezanezhad 2025-04-30 16:01:52 +02:00
  • 208bde706f resolving issue #158 vahidrezanezhad 2025-04-30 13:55:09 +02:00
  • 3e8adb86c2
    Merge pull request #157 from qurator-spk/kba-patch-1 Konstantin Baierer 2025-04-29 11:42:18 +02:00
  • 77dae129d5
    CI: Use most recent actions/setup-python@v5 Konstantin Baierer 2025-04-22 13:22:28 +02:00
  • 192b9111e3 updating eynollah README, how to use it for use cases vahidrezanezhad 2025-04-22 00:23:01 +02:00
  • b4df978dd5
    Merge pull request #154 from qurator-spk/ci-pypi Clemens Neudecker 2025-04-17 17:01:20 +02:00
  • 30ba234641 CI: pypi kba 2025-04-16 19:27:17 +02:00
  • 41318f0404 📝 changelog kba 2025-04-15 11:14:26 +02:00
  • a22df11ebb Restoring the contour in the original image caused an error due to an empty tuple. This issue has been resolved, and as expected, the confidence score for this contour is set to zero vahidrezanezhad 2025-04-14 00:42:08 +02:00
  • 3b1886973d gha snippets from codecov sample page test-codecov kba 2025-04-08 14:52:38 +02:00
  • 60a05711bb test kba 2025-04-08 14:46:22 +02:00
  • 8080bd823c 📦 v0.4.0 v0.4.0 kba 2025-04-07 16:48:57 +02:00
  • bcf1898aa4 📝 changelog Robert Sachunsky 2025-04-07 16:46:58 +02:00
  • 177e017167 test_run: ensure exceptions are shown Robert Sachunsky 2025-04-06 18:24:56 +00:00
  • 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. vahidrezanezhad 2025-04-06 00:33:36 +02:00
  • 132d3e3d27 CI: use clash-free artifact name for report upload Robert Sachunsky 2025-04-05 11:36:21 +02:00
  • dc64079b6b CI: fix coverage report calls Robert Sachunsky 2025-04-05 03:40:02 +02:00
  • 7609c64c8b CI: make coverage cfg work with both editable and dist install Robert Sachunsky 2025-04-05 03:05:26 +02:00
  • bbc06dbbc1 CI: forgot to (re-)enable verbose logging Robert Sachunsky 2025-04-05 02:10:52 +02:00
  • a41f18b13d CI: (try to) store/upload coverage results Robert Sachunsky 2025-04-05 01:34:28 +02:00
  • 4339444e47 binarization CLI: fix option checks, simplify to asserts, fix dir_in mode Robert Sachunsky 2025-04-05 01:21:08 +02:00