mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-12-13 16:09:04 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity
4,237 Commits 10 Branches 66 Tags 67 MiB
88174fea1d
Commit Graph

75 Commits

Author SHA1 Message Date
Shree
6a671e39df remove legacy parameter disable_character_fragments from lstm.train 2019-11-01 15:08:52 +01:00
zdenop
b244bd1c13 Removed unused parameters 
The following parameters are not used anywhere anymore:

 * use_definite_ambigs_for_classifier
 * max_viterbi_list_size
 * word_to_debug_lengths
 * fragments_debug
 * tessedit_redo_xheight
 * debug_acceptable_wds
 * tessedit_matcher_log
 * tessedit_test_adaption_mode
 * docqual_excuse_outline_errs
 * crunch_pot_garbage
 * suspect_space_level
 * tessedit_consistent_reps
 * wordrec_display_all_words
 * wordrec_no_block
 * wordrec_worst_state
 * fragments_guide_chopper
 * segment_adjust_debug
 * classify_adapt_feature_thresh (classify_adapt_feature_threshold still exists)
 * classify_adapt_proto_thresh (classify_adapt_proto_threshold still exists)
 * classify_min_norm_scale_x
 * classify_max_norm_scale_x
 * classify_min_norm_scale_y
 * classify_max_norm_scale_y
 * il1_adaption_test
 * textord_blob_size_bigile
 * textord_blob_size_smallile
 * editor_debug_config_file
 * textord_tabfind_show_color_fit

The list was generated by a python script and each parameter occurence checked
manually.

# Conflicts:
#	src/classify/classify.cpp
 2019-11-01 14:58:36 +01:00
Julian Gilbey
d8f695271f fix #2616: allow building of training data 
This fixes Issue #2616 by preventing an attempt to build the recognition engine when running tesstrain.sh.
 2019-11-01 13:59:26 +01:00
Stefan Weil
7db25e15c0 Remove unused config variable tessedit_single_match 
Replace also TRUE, FALSE by true, false.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2019-03-31 17:38:35 +02:00
Shree
08e96516c6 install lstmbox and wordstrbox config files 2019-03-01 15:26:59 +00:00
Shree Devi Kumar
f3362a4b5b Add renderer to create WordStr box files from images 2019-02-10 19:59:17 +00:00
zdenop
2ae65b2493
Merge pull request #2216 from Shreeshrii/lstmbox 
Lstmbox
 2019-02-10 13:53:41 +01:00
Chris Mayo
6dc48adfee Rename get.image config to get.images and install 2019-02-05 19:57:53 +00:00
Shree Devi Kumar
9c89cd51cf Add a new renderer to create box files from images for LSTM training 
(cherry picked from commit 921da6be2bdbda2ddd64514f9b6bec40a336246a)

fix typo

(cherry picked from commit 7bd1a0c80393fce2f34e2845cb26760bcf3791cd)

Add lstmboxrenderer to CMakeLists

(cherry picked from commit cfef3a889aef830725921b5c0218d5e9c633b03e)

fix formatting

(cherry picked from commit 7ba2b01ede7940ed609a073364948ef8c838cd10)
 2019-02-05 14:03:29 +00:00
Stefan Weil
e817d93e62 Add configuration file for ALTO to installation 
Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-11-30 06:17:04 +01:00
Jake Sebright
d7cee03a94 Add support for ALTO output 2018-11-30 06:09:36 +01:00
Zdenko Podobný
ba64aaf257 add lstmdebug config to distribution and installation process 2018-10-29 09:38:11 +01:00
Stefan Weil
125fdc3f1b Add debug configuration for LSTM 
It was provided by Jeff Breidenbach <jbreiden@google.com>.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-10-27 08:04:45 +02:00
Zdenko Podobný
3d508a65a7 set unlv_tilde_crunching to false; fixes #1449 #948 2018-10-23 09:26:32 +02:00
Stefan Weil
c6f759148b Don't set page segmentation mode for unlv config 
Setting the page segmentation mode to 6 ("Assume a single uniform block
of text") typically improves the layout detection for such texts, but
should not be done in the config file.

unlvtests/runtestset.sh adds `--psm 6` explicitly, so test results
won't change when using that script.

This is similar to commit ecfee53bac.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-10-04 21:01:18 +02:00
Stefan Weil
ecfee53bac Don't set page segmentation mode for hocr, pdf and tsv configs 
Setting the page segmentation mode in those config files gives unexpected
results: the text recognized when no config or only txt is given changes
if both txt and any of hocr, pdf or tsv is chosen.

In a test set of nearly 200 pages from historical books, using
segmentation mode 1 is typically slightly better than the default,
but there are also cases where it is much worse. Therefore the user
should be able to decide which page segmentation mode is best.

Old results for hocr, pdf or tsv now need an explicit `--psm 1` for
reproduction.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-10-04 12:05:49 +02:00
Stefan Weil
dabf3c299f Fix file endings 
Text files should end with a LF, but not additional empty lines.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-04-25 19:35:33 +02:00
Stefan Weil
10a8a67ca2 Remove execute permission from config file (#1263) 
This fixes the only configuration file which had such permissions.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
 2018-01-10 16:43:02 +01:00
Atsuyoshi Suzuki
82d62f89a2 Update Makefile.am (add 'lstm.train') 2017-04-02 17:06:12 +09:00
Ray Smith
65517794f9 Added missing lstm.train 2016-12-06 08:48:23 -08:00
Ray Smith
2c837dffc3 Result of clang tidy on recent merge 2016-11-07 10:46:33 -08:00
Tom Morris
fc80ceafb9 Fix hocrtsv references in Makefile 2016-03-02 10:46:52 -05:00
Tom Morris
6700edd8bc Cleanup TSV renderer 
Remove all references to hocr, hocr.tsv, etc. Remove dead code for font
info, input filename, HTML escapes. Improved comments. Fixed
indentation.
 2016-03-01 13:41:19 -05:00
Sundar M. Vaidya
937ceb2d1b Adds hocrtsv to tessdata/configs/Makefile.am 2016-03-01 12:25:15 -05:00
Sundar M. Vaidya
3163b38151 Adds hocrtsv file to configs folder. 2016-03-01 12:23:12 -05:00
Sundar M. Vaidya
59d593d796 Calls TessHOcrTsvRenderer if tessedit_create_hocrtsv is true. 2016-03-01 12:23:12 -05:00
Tom Morris
e3e1fe0e20 Document hocr_font_info in config 2016-02-14 16:49:00 -05:00
Amit Dovev
6b08184a2c Update Makefile.am 2015-12-18 16:12:32 +02:00
amitdo
c2f5e9b849 If there is no explicit renderer(s), default to TessTextRenderer 
Revert fd429c32, 43834da7, 05de195e.

See #49, #59.

The code in this commit solves the issue in a more elegant way, IMHO.

Now you can use:
  * `tesseract eurotext.tif eurotext txt pdf`
  * `tesseract eurotext.tif eurotext txt hocr`
  * `tesseract eurotext.tif eurotext txt hocr pdf`

NOTE:
  With `tesseract eurotext.tif eurotext`
  or `tesseract eurotext.tif eurotext txt`
  the psm will be set to '3', but...
  With `tesseract eurotext.tif eurotext txt pdf`
  or `tesseract eurotext.tif eurotext txt hocr`
  the psm will be set to '1'.
 2015-12-11 19:06:49 +02:00
Zdenko Podobný
66a76a9477 Revert "temporary add config/*, configure and Makefile.in for release" 
This reverts commits ec9581d8f2, 1afe382c4e, 4b2cfabcc1
 2015-07-31 21:44:43 +02:00
Zdenko Podobný
5dfb0cb898 Fixes #64 - tessedit_create_txt 0 blocks box training 2015-07-25 22:49:55 +02:00
Jim O'Regan
05de195efc disable text creation for unlv, makebox, box.train, and box.train.stderr (see #49) 2015-07-20 10:07:55 +01:00
Jim O'Regan
43834da7a2 disable text creation when creating hOCR (issue #49) 2015-07-18 08:56:21 +01:00
Jeff Breidenbach
fd429c32a0 PDF creation: not disabling tessedit_create_txt 
Okay, everything is more of less under control except for this:

  tesseract phototest.tif - pdf > phototest.pdf

This is sending activating both the text renderer, and the pdf renderer.
They both get sent to stdout where they mix together and cause chaos.
Same thing happens with this command.

   tesseract phototest.tif stdout pdf > phototest.pdf

What's happening is tesseractmain.cpp is setting tessedit_create_pdf without
disabling tessedit_create_txt.

https://groups.google.com/d/msgid/tesseract-dev/32c065ee-aefa-441a-b37b-b6bdc234c8ab%40googlegroups.com
 2015-07-18 08:39:57 +01:00
Zdenko Podobný
ec9581d8f2 temporary add configure and Makefile.in for release 2015-07-11 09:42:43 +02:00
Zdenko Podobný
369fabb7fc fix filemode; 
update autotools and distribution script to repository changes;
ignore doxygen generated files and langauge data files;
 2014-08-14 23:37:17 +02:00
theraysmith@gmail.com
91d2265429 More minor fixes from issues and cleanup 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@974 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2014-01-10 01:38:00 +00:00
theraysmith@gmail.com
4c72deea6c Added pdf config file 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@972 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2014-01-09 19:18:07 +00:00
zdenop@gmail.com
53a3e0f88a fix issue 755; add example config files from tesseract manpage 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@894 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2013-10-20 20:20:10 +00:00
zdenop@gmail.com
32d212d0c6 add new config file - get.image 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@826 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2013-02-23 11:56:49 +00:00
zdenop@gmail.com
e83503022c update script for 3.02.02 release 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@793 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2012-10-26 18:49:14 +00:00
zdenop@gmail.com
3b326532cc fix --enable-multiple-libraries; implement quite mode (issue 580) 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@691 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2012-03-03 11:48:59 +00:00
theraysmith@gmail.com
d581ab7e12 New config for testing bigram correction. 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@661 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2012-02-02 18:46:19 +00:00
theraysmith@gmail.com
6e273b71bd Cube trained data for fra, ita, rus, spa 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@656 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2012-02-02 03:08:26 +00:00
joregan@gmail.com
323ee5af7a more Makefile.in 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@618 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2011-08-18 18:40:33 +00:00
theraysmith@gmail.com
d5d15f32d7 Deleted Makefile.in from svn 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@606 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2011-08-18 16:32:44 +00:00
zdenop@gmail.com
3463abfd34 commented parameters that caused error (read_params_file: parameter not found:) 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@589 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2011-06-15 20:20:45 +00:00
theraysmith
311d1f9253 Added Hindi traineddata 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@576 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2011-03-21 21:57:08 +00:00
theraysmith
5c854e03ea Cleaned up unused parameters 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@539 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2010-11-30 01:06:44 +00:00
zdenop@gmail.com
4523ce9f7d 3.01 code from http://github.com/jimregan/tesseract-ocr with addaptions related to Linux and Windows (VC2008) compile process 
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@526 d0cd1f9f-072b-0410-8dd7-cf729c803f20
 2010-11-23 18:34:14 +00:00