tesseract/doc/wordlist2dawg.1.asc
david.eger@gmail.com 58e06c8c45 Update man pages for Tesseract 3.02.
git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@670 d0cd1f9f-072b-0410-8dd7-cf729c803f20
2012-02-09 22:55:47 +00:00

70 lines
1.5 KiB
Plaintext

WORDLIST2DAWG(1)
================
:doctype: manpage
NAME
----
wordlist2dawg - convert a wordlist to a DAWG for Tesseract
SYNOPSIS
--------
*wordlist2dawg* 'WORDLIST' 'DAWG' 'lang.unicharset'
*wordlist2dawg* -t 'WORDLIST' 'DAWG' 'lang.unicharset'
*wordlist2dawg* -r 1 'WORDLIST' 'DAWG' 'lang.unicharset'
*wordlist2dawg* -r 2 'WORDLIST' 'DAWG' 'lang.unicharset'
*wordlist2dawg* -l <short> <long> 'WORDLIST' 'DAWG' 'lang.unicharset'
DESCRIPTION
-----------
wordlist2dawg(1) converts a wordlist to a Directed Acyclic Word Graph
(DAWG) for use with Tesseract. A DAWG is a compressed, space and time
efficient representation of a word list.
OPTIONS
-------
-t
Verify that a given dawg file is equivalent to a given wordlist.
-r 1
Reverse a word if it contains an RTL character.
-r 2
Reverse all words.
-l <short> <long>
Produce a file with several dawgs in it, one each for words
of length <short>, <short+1>,... <long>
ARGUMENTS
---------
'WORDLIST'
A plain text file in UTF-8, one word per line.
'DAWG'
The output DAWG to write.
'lang.unicharset'
The unicharset of the language. This is the unicharset
generated by mftraining(1).
SEE ALSO
--------
tesseract(1), combine_tessdata(1), dawg2wordlist(1)
<http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract3>
COPYING
-------
Copyright \(C) 2006 Google, Inc.
Licensed under the Apache License, Version 2.0
AUTHOR
------
The Tesseract OCR engine was written by Ray Smith and his research groups
at Hewlett Packard (1985-1995) and Google (2006-present).