Lluis Gomez-Bigorda
|
c3042c3f51
|
remove trailing space
|
2015-08-26 21:47:36 +02:00 |
|
Lluis Gomez-Bigorda
|
a94753c0bf
|
fix Win x64 warning
|
2015-08-26 21:17:52 +02:00 |
|
Lluis Gomez-Bigorda
|
b19c676559
|
fix Win x64 and iOS warnings
|
2015-08-26 21:04:27 +02:00 |
|
Lluis Gomez-Bigorda
|
7cd6dbb0c9
|
Adds two more examples of croped word recognition, and show how to set the beam size parameter for small lexicon.
|
2015-08-26 20:13:15 +02:00 |
|
Lluis Gomez-Bigorda
|
cc43177675
|
OCRBeamSearchDecoder code refactor: fixes most of the TODOs comments, cleans-up code, and fix a bug in score_segmentation(...) member function.
|
2015-08-26 20:11:17 +02:00 |
|
Lluis Gomez-Bigorda
|
0be529da05
|
Adds demo program of Scene Text Character Recognition using the CNN classifier.
|
2015-08-21 18:20:41 +02:00 |
|
Lluis Gomez-Bigorda
|
796c26f68a
|
Fix w64 warnings
|
2015-08-03 22:15:31 +02:00 |
|
Lluis Gomez-Bigorda
|
1481194b54
|
Fix w64 warnings
|
2015-08-03 21:45:45 +02:00 |
|
Lluis Gomez-Bigorda
|
2315f9ca22
|
Minor bugfix: removes unwanted space character at the begining of recognition output strings.
|
2015-08-03 20:53:49 +02:00 |
|
Lluis Gomez-Bigorda
|
2538bf74a6
|
Adds example on segmented word recognition. Shows the use of the OCRHMMDecoder with the NM and CNN default classifiers.
|
2015-08-03 20:50:33 +02:00 |
|
lluis
|
ee677a255b
|
Overload the run() method in BaseOCR class in order to adapt to different classifier callbacks. The original run() method accepts only one Mat input image, this is expected to be a binarzed image with black and white text and works both with the OCRTesseract class and the OCRHMMDecoder class when the character classifier callback works with binary images (e.g. NM). The new run() method accepts two Mat input parameters. One for the gray scale (or color) source image and the other for a binary mask where each connected component corresponds to a pre-segmented character in the input image. This way the OCRHMMDecoder is able to work with character classifiers that operate in grey scale (or color) images (e.g. a CNN).
|
2015-08-01 13:52:24 +02:00 |
|
lluis
|
c146d37b93
|
Better CNN model for character recognition. Trained with an augmented dataset by adding translation/scale variations. Updated the croped word recognition with new class numbering (compatible with previous NM classifier).
|
2015-08-01 13:52:00 +02:00 |
|
Vadim Pisarevsky
|
f9d42886f1
|
Merge pull request #298 from lluisgomez:master
|
2015-07-31 09:44:46 +00:00 |
|
Vadim Pisarevsky
|
f9969b0da6
|
Merge pull request #308 from lluisgomez/4373
Fix bug #4373:
|
2015-07-24 14:18:20 +03:00 |
|
lluis
|
43b5febc63
|
Fix bug #4373: Error (Assertion failed in resize) when passing very elongated contours to the recognition module
|
2015-07-24 11:39:42 +02:00 |
|
lluis
|
015e00bcdb
|
Fix memory leak bug #4420
|
2015-07-23 20:01:19 +02:00 |
|
lluis
|
2e787d6b42
|
fix Winx64 warnings
|
2015-07-09 19:13:42 +02:00 |
|
lluis
|
00f670e779
|
update to use the new createOCRHMMTransitionsTable() function, and fix program description in header comments
|
2015-07-09 18:46:51 +02:00 |
|
lluis
|
e4cc9a77a9
|
Adds createOCRHMMTransitionsTable() utility function to create a tailored language model transitions table from a given list of words (lexicon)
|
2015-07-09 18:40:11 +02:00 |
|
lluis
|
c5bfcb95d1
|
fix doxygen warnings
|
2015-06-26 14:27:52 +02:00 |
|
lluis
|
7940d4d92e
|
fix win64 compilation error: arrays must be defined with compile-time fixed size :)
|
2015-06-26 14:17:29 +02:00 |
|
lluis
|
a6b46f8b6a
|
fix compilation warnings
|
2015-06-26 13:20:21 +02:00 |
|
lluis
|
662e2ce561
|
trailing whitespaces
|
2015-06-26 13:15:28 +02:00 |
|
lluis
|
9c55b98813
|
Add a demo program for the OCRBeamSearchDecoder class and needed data files
|
2015-06-26 13:04:33 +02:00 |
|
lluis
|
52cca0ddc6
|
Adds a first implementation of the OCRBeamSearchDecoder class using the Single Layer CNN character classifier described in Coates, Adam, et al. paper: Text detection and character recognition in scene images with unsupervised feature learning, ICDAR 2011
|
2015-06-26 13:03:50 +02:00 |
|
Lluis Gomez-Bigorda
|
d7410500e0
|
Fix for Bug #3633: do away with "quads [2][3] = 255;" The four lowest bits in each quads[i][j] correspond to the 2x2 binary patterns Q_1, Q_2, Q_3 in the NM paper [1] (see in page 4 at the end of first column). Q_1 and Q_2 have four patterns, while Q_3 has only two.
|
2015-06-09 16:24:36 +02:00 |
|
lluis
|
2dc4cd934b
|
Fix for Bug 4074. This seems to be just a typo-error, because the Tesseract API can handle correctly with RGB images (double-checked and it works).
|
2015-06-09 09:36:14 +02:00 |
|
Vadim Pisarevsky
|
cf3dff2e9a
|
fixed several warnings in opencv_contrib (text, xfeatures2d); added tesseract installation mini-guide in text/readme.md
|
2015-06-02 01:16:03 +03:00 |
|
Vadim Pisarevsky
|
ec2b3ed662
|
Merge pull request #185 from mshabunin/remove-docs
Removed sphinx documentation
|
2015-03-16 13:07:31 +03:00 |
|
Maksim Shabunin
|
67a2066c4b
|
Java and python wrappers for contrib modules
|
2015-02-26 15:46:02 +03:00 |
|
Maksim Shabunin
|
7d9bbdcaad
|
Remove all sphinx files
|
2015-02-17 12:49:37 +03:00 |
|
Maksim Shabunin
|
2b74ca6f12
|
Added opencv_contrib_world module and fixed some compilation warnings on iOS
|
2015-02-10 18:27:52 +03:00 |
|
Vadim Pisarevsky
|
c6ea683b86
|
Merge pull request #155 from nisargthakkar/Bug4074
Fix for Bug 4074. Both conditions of if are same
|
2015-01-15 15:02:29 +03:00 |
|
Jay Bosamiya
|
e2253ea471
|
Fix typo
|
2015-01-13 18:19:47 +05:30 |
|
Jay Bosamiya
|
58f2dc3b4c
|
Fix typo
|
2015-01-13 18:19:08 +05:30 |
|
Nisarg Thakkar
|
c21fd589db
|
Fix for Bug 4074. Both conditions of if are same
|
2015-01-08 12:24:19 +05:30 |
|
Maksim Shabunin
|
a20c5c8dd9
|
Doxygen documentation for all modules
|
2014-12-01 15:54:42 +03:00 |
|
Vadim Pisarevsky
|
08699f115e
|
Merge pull request #106 from mtamburrano/adas_fix
adas fix
|
2014-11-07 18:18:46 +03:00 |
|
Vadim Pisarevsky
|
38bdd6f3f4
|
Merge pull request #110 from StevenPuttemans/fix_contrib_README
adding more clear documentation
|
2014-11-07 18:16:45 +03:00 |
|
Maksim Shabunin
|
bad3cd2427
|
Fixed some existing doxygen comments
|
2014-10-21 16:03:02 +04:00 |
|
manuele
|
7f97464cc3
|
logtopolar
|
2014-10-21 12:33:40 +02:00 |
|
Sancho McCann
|
6d4c23a63a
|
Corrected function call.
|
2014-10-19 23:20:17 -07:00 |
|
Sancho McCann
|
f83cffe74a
|
Bugfix: NNs not found due to precision error.
|
2014-10-19 09:59:56 -07:00 |
|
Vadim Pisarevsky
|
218358a5d5
|
corrected text detection once again to match the new MSER interface
|
2014-10-16 22:00:19 +04:00 |
|
Vadim Pisarevsky
|
0cfd795303
|
refactored xfeatures2d in the same style as features2d
|
2014-10-16 16:33:21 +04:00 |
|
StevenPuttemans
|
c20ea8e7e8
|
adding more clear documentation
|
2014-10-10 21:17:46 +02:00 |
|
Vadim Pisarevsky
|
071eb5a7c5
|
Merge pull request #69 from thorikawa/fix-usage-text
Fix key name to switch between MSER/CSER regions.
|
2014-08-27 14:15:40 +04:00 |
|
Vadim Pisarevsky
|
e4cfe59a99
|
Merge pull request #74 from znah/to_python
[WIP] Effort to expose new contributions into python
|
2014-08-25 22:19:40 +04:00 |
|
Sancho McCann
|
86dd8f745a
|
Bugfix: Precision error.
When p >= 1, setting p to 0.999999999999999999999999999999, didn't actually make p less than one.
|
2014-08-22 11:25:44 -07:00 |
|
Alexander Mordvintsev
|
3f7cca4a6f
|
fixed 'xphoto' and 'text' build on MSVC 2008
|
2014-08-22 14:05:45 +04:00 |
|