Commit Graph

20 Commits

Author SHA1 Message Date
Zdenko Podobný
9d1db80c69 backport from 4.00: changes from ccstruct excluding imagedata 2016-11-30 19:29:59 +01:00
Zdenko Podobn??
6234da6686 backport from 4.00: changes in wordrec + FakeWordFromRatings 2016-11-30 10:56:25 +01:00
Zdenko Podobn??
70ed782ab5 style fix 2016-11-30 10:38:08 +01:00
Zdenko Podobný
90651e111f backport style changes from 4.00 for better identification of fixes and new code 2016-11-25 15:14:46 +01:00
Stefan Weil
4a92ff5862 Fix compiler warnings for copy constructors
gcc reports these warnings with -Wextra:

ccstruct/pageres.h:330:3: warning:
 base class 'class ELIST_LINK' should be explicitly initialized
 in the copy constructor [-Wextra]
ccstruct/ratngs.cpp:115:1: warning:
 base class 'class ELIST_LINK' should be explicitly initialized
 in the copy constructor [-Wextra]
ccstruct/ratngs.h:291:3: warning:
 base class 'class ELIST_LINK' should be explicitly initialized
 in the copy constructor [-Wextra]
ccutil/genericvector.h:435:3: warning:
 base class 'class GenericVector<WERD_RES*>' should be explicitly initialized
 in the copy constructor [-Wextra]

Signed-off-by: Stefan Weil <sw@weilnetz.de>
2015-11-05 09:19:37 +01:00
Stefan Weil
edf765b952 Remove unneeded const qualifiers
This fixes compiler warnings like this one:

api/baseapi.h:739:32: warning:
 type qualifiers ignored on function return type [-Wignored-qualifiers]

Signed-off-by: Stefan Weil <sw@weilnetz.de>
2015-11-05 06:36:42 +01:00
Ray Smith
84920b92b3 Font and classifier output structure cleanup.
Font recognition was poor, due to forcing a 1st and 2nd choice at
a character level, when the total score for the correct font is often
correct at the word level, so allowed the propagation of a full set
of fonts and scores to the word recognizer, which can now decide word
level fonts using the scores instead of simple votes.

Change precipitated a cleanup of output data structures for classifier
results, eliminating ScoredClass and INT_RESULT_STRUCT, with a few
extra elements going in UnicharRating, and using that wherever possible.
That added the extra complexity of 1-rating due to a flip between 0 is
good and 0 is bad for the internal classifier scores before they are
converted to rating and certainty.
2015-05-12 17:24:34 -07:00
Ray Smith
0e868ef377 Major change to improve layout analysis for heavily diacritic languages:
Tha, Vie, Kan, Tel etc.
There is a new overlap detector that detects when diacritics
cause a big increase in textline overlap. In such cases, diacritics from
overlap regions are kept separate from layout analysis completely, allowing
textline formation to happen without them. The diacritics are then assigned
to 0, 1 or 2 close words at the end of layout analysis, using and modifying
an old noise detection data path.
The stored diacritics are used or not during recognition according to the
character classifier's liking for them.
2015-05-12 16:47:02 -07:00
Ray Smith
b6d0184806 Fixed problems with shifted baselines so recognition can recover from layout analysis errors. 2015-05-12 15:53:45 -07:00
theraysmith@gmail.com
dbf6197471 Major refactor of control.cpp to enable line recognition
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@1147 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2014-08-11 23:23:06 +00:00
theraysmith@gmail.com
7ec4fd7a56 Refactorerd control functions to enable parallel blob classification
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@904 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2013-11-08 20:30:56 +00:00
theraysmith@gmail.com
4d514d5a60 Major refactor of beam search, elimination of dead code, misc bug fixes, updates to Makefile.am, Changelog etc.
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@878 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2013-09-23 15:26:50 +00:00
theraysmith@gmail.com
9206e92b0d Added simultaneous multi-language capability, Refactored top-level word recognition module, Blamer module added for error analysis, Tidied up constraints on control parameters, Added UNICHARSET to WERD_CHOICE to make mult-language handling easier, Added word bigram correction
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@655 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2012-02-02 03:06:39 +00:00
theraysmith
82b1b201fc Various fixes, including memory leak in fixspace, font labels on output, removed some annoying debug output, fixes to initialization of parameters, general cleanup, and added Hindi
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@568 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2011-03-21 21:44:45 +00:00
theraysmith
5a779704da Deleted lots of dead code, including PBLOB
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@557 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2011-03-18 21:52:38 +00:00
theraysmith
137f4806b6 Added sub/superscript, small/dropcap detection
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@547 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2010-12-09 01:32:20 +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
theraysmith
51ed03368d Fixes to lists so an empty constructor is not needed + reenable debugging
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@207 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2008-12-30 18:15:44 +00:00
theraysmith
c4f4840fbe Fixed name collision with jpeg library
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@163 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2008-04-22 00:41:37 +00:00
tmbdev
425d593ebe top-skimming import from sf.net
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk/trunk@2 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2007-03-07 20:03:40 +00:00