vahidrezanezhad
|
4de441eaaa
|
OCR prediction is now enabled to integrate results from both RGB and binarized images or to be performed on each individually
|
2 weeks ago |
|
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
|
2 weeks ago |
|
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.
|
2 weeks ago |
|
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.
|
2 weeks ago |
|
vahidrezanezhad
|
b55389ac62
|
Update requirements.txt
|
2 weeks ago |
|
vahidrezanezhad
|
8bf70d905f
|
Merge pull request #147 from qurator-spk/revert-146-remove-imutils-1
Revert "replace usages of `imutils` with opencv equivalents"
|
2 weeks ago |
|
vahidrezanezhad
|
f756b08c9b
|
Revert "replace usages of `imutils` with opencv equivalents"
|
2 weeks ago |
|
vahidrezanezhad
|
c9de578d4d
|
removing imutils from requirements
|
2 weeks ago |
|
vahidrezanezhad
|
52c605185a
|
Merge pull request #146 from qurator-spk/remove-imutils-1
replace usages of `imutils` with opencv equivalents
|
2 weeks ago |
cneud
|
0e9a72ea52
|
consolidate usage documentation
|
2 weeks ago |
|
cneud
|
3a55b6ce91
|
consolidate usage documentation
|
2 weeks ago |
|
cneud
|
e9fa691308
|
add model and training documentation
|
2 weeks ago |
|
vahidrezanezhad
|
6f36c7177f
|
For OCR, the splitting ratio of text lines is adjusted
|
2 weeks ago |
|
cneud
|
181c0c584f
|
bbox rotation with opencv
|
2 weeks ago |
|
cneud
|
eaff9e3537
|
Merge branch 'main' into remove-imutils-1
|
2 weeks ago |
|
vahidrezanezhad
|
7df0427b04
|
In the context of OCR, if Page-XML files already contain text, the new predicted text will replace the existing text.
|
2 weeks ago |
|
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.
|
2 weeks ago |
|
Clemens Neudecker
|
005b6988f4
|
Merge pull request #140 from qurator-spk/machine_based_reading_order_integration
Machine based reading order integration
|
3 weeks ago |
|
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.
|
3 weeks ago |
|
vahidrezanezhad
|
c8b8529951
|
For the CNN-RNN OCR model, long text lines are split into two segments
|
4 weeks ago |
|
vahidrezanezhad
|
aa72ca3006
|
Resolved an issue in the OCR-D framework where dir_out received a None value
|
1 month ago |
|
vahidrezanezhad
|
a4f1f35125
|
Resolving test failure
|
1 month ago |
kba
|
54040c1db4
|
Merge remote-tracking branch 'bertsky/machine_based_reading_order_integration_fixes' into machine_based_reading_order_integration
|
1 month ago |
|
cneud
|
0b2c1b9275
|
remove `imutils` dependency
|
1 month ago |
|
Clemens Neudecker
|
687aba1fa2
|
replace usages of `imutils` with opencv equivalents
should fix https://github.com/qurator-spk/eynollah/issues/141
|
1 month ago |
|
vahidrezanezhad
|
7110bd971f
|
resolved an error for light version in the case that slope_deskew is smaller than slope_threshold
|
1 month ago |
|
vahidrezanezhad
|
25116a2c79
|
resolved 2 errors
|
2 months ago |
|
vahidrezanezhad
|
33fda2f8be
|
changing cnn ocr model name
|
4 months ago |
Robert Sachunsky
|
335aa273a1
|
simplify, wrap extremely long lines
|
4 months ago |
|
Robert Sachunsky
|
cfc65128b1
|
reduce redundancy/indentation
|
4 months ago |
|
Robert Sachunsky
|
01376af905
|
do_order_of_regions_with_model: simplify
|
4 months ago |
|
vahidrezanezhad
|
92bfac4b41
|
Provide OCR as an option to process a directory of XML files, incorporating layout and text line coordinates.
|
4 months ago |
|
vahidrezanezhad
|
fbeef79d50
|
adding scatter_nd inference
|
4 months ago |
|
Robert Sachunsky
|
0ae28f7d3e
|
switch from stdlib to loky.ProcessPoolExecutor, ensure shutdown
|
4 months ago |
|
vahidrezanezhad
|
f93c6c288d
|
function of patch-wise inference with scatter_nd is added
|
4 months ago |
|
vahidrezanezhad
|
0e8c561618
|
debugging issues
|
4 months ago |
|
Robert Sachunsky
|
e9c0d716f6
|
CI: install optional dependencies, too
|
4 months ago |
|
Robert Sachunsky
|
dcaf796283
|
change polarity of orientation angle (PAGE schema required cw=positive)
|
4 months ago |
|
Robert Sachunsky
|
b4b0890294
|
add option to overwrite output xml, but skip by default if file exists
|
4 months ago |
|
Robert Sachunsky
|
b9ca7a6191
|
log num_cols-dependent resizing
|
4 months ago |
|
Robert Sachunsky
|
9270ea4550
|
annotate region angles in PAGE
|
4 months ago |
|
Robert Sachunsky
|
3b70b11ea6
|
avoid deskewing patches if binary-empty
|
4 months ago |
|
Robert Sachunsky
|
7e9ee90e6e
|
switch from (ad-hoc) mp.Pool to (attribute) concurrent.futures.ProcessPoolExecutor
|
4 months ago |
|
Robert Sachunsky
|
68456ea002
|
do_work_of_slopes_new*, do_back_rotation_and_get_cnt_back, do_work_of_contours_in_image: use mp.Pool, simplify
|
4 months ago |
|
Robert Sachunsky
|
25e967397d
|
exit early if no text regions found (to avoid segfault)
|
4 months ago |
|
Robert Sachunsky
|
21efea8711
|
no del on function argument
|
4 months ago |
|
Robert Sachunsky
|
5e0c1da711
|
simplify
|
4 months ago |
|
Robert Sachunsky
|
54cb15056b
|
do_image_rotation / return_deskew_slop: avoid code duplication, simplify via mp.Pool
|
4 months ago |
|
Robert Sachunsky
|
6fe02df973
|
do_image_rotation: fix f93fa12 (do return results)
|
4 months ago |
|
Robert Sachunsky
|
d68017037c
|
do_prediction: trigger GC to avoid CUDA OOM
|
4 months ago |