tesseract/doc/combine_tessdata.1.xml
2015-06-13 00:08:05 +02:00

282 lines
8.8 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN" "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd">
<?asciidoc-toc?>
<?asciidoc-numbered?>
<refentry lang="en">
<refentryinfo>
<title>COMBINE_TESSDATA(1)</title>
</refentryinfo>
<refmeta>
<refentrytitle>combine_tessdata</refentrytitle>
<manvolnum>1</manvolnum>
<refmiscinfo class="source">&#160;</refmiscinfo>
<refmiscinfo class="manual">&#160;</refmiscinfo>
</refmeta>
<refnamediv>
<refname>combine_tessdata</refname>
<refpurpose>combine/extract/overwrite Tesseract data</refpurpose>
</refnamediv>
<refsynopsisdiv id="_synopsis">
<simpara><emphasis role="strong">combine_tessdata</emphasis> [<emphasis>OPTION</emphasis>] <emphasis>FILE</emphasis>&#8230;</simpara>
</refsynopsisdiv>
<refsect1 id="_description">
<title>DESCRIPTION</title>
<simpara>combine_tessdata(1) is the main program to combine/extract/overwrite
tessdata components in [lang].traineddata files.</simpara>
<simpara>To combine all the individual tessdata components (unicharset, DAWGs,
classifier templates, ambiguities, language configs) located at, say,
/home/$USER/temp/eng.* run:</simpara>
<literallayout class="monospaced">combine_tessdata /home/$USER/temp/eng.</literallayout>
<simpara>The result will be a combined tessdata file /home/$USER/temp/eng.traineddata</simpara>
<simpara>Specify option -e if you would like to extract individual components
from a combined traineddata file. For example, to extract language config
file and the unicharset from tessdata/eng.traineddata run:</simpara>
<literallayout class="monospaced">combine_tessdata -e tessdata/eng.traineddata \
/home/$USER/temp/eng.config /home/$USER/temp/eng.unicharset</literallayout>
<simpara>The desired config file and unicharset will be written to
/home/$USER/temp/eng.config /home/$USER/temp/eng.unicharset</simpara>
<simpara>Specify option -o to overwrite individual components of the given
[lang].traineddata file. For example, to overwrite language config
and unichar ambiguities files in tessdata/eng.traineddata use:</simpara>
<literallayout class="monospaced">combine_tessdata -o tessdata/eng.traineddata \
/home/$USER/temp/eng.config /home/$USER/temp/eng.unicharambigs</literallayout>
<simpara>As a result, tessdata/eng.traineddata will contain the new language config
and unichar ambigs, plus all the original DAWGs, classifier templates, etc.</simpara>
<simpara>Note: the file names of the files to extract to and to overwrite from should
have the appropriate file suffixes (extensions) indicating their tessdata
component type (.unicharset for the unicharset, .unicharambigs for unichar
ambigs, etc). See k*FileSuffix variable in ccutil/tessdatamanager.h.</simpara>
<simpara>Specify option -u to unpack all the components to the specified path:</simpara>
<literallayout class="monospaced">combine_tessdata -u tessdata/eng.traineddata /home/$USER/temp/eng.</literallayout>
<simpara>This will create /home/$USER/temp/eng.* files with individual tessdata
components from tessdata/eng.traineddata.</simpara>
</refsect1>
<refsect1 id="_options">
<title>OPTIONS</title>
<simpara><emphasis role="strong">-e</emphasis> <emphasis>.traineddata</emphasis> <emphasis>FILE</emphasis>&#8230;:
Extracts the specified components from the .traineddata file</simpara>
<simpara><emphasis role="strong">-o</emphasis> <emphasis>.traineddata</emphasis> <emphasis>FILE</emphasis>&#8230;:
Overwrites the specified components of the .traineddata file
with those provided on the comand line.</simpara>
<simpara><emphasis role="strong">-u</emphasis> <emphasis>.traineddata</emphasis> <emphasis>PATHPREFIX</emphasis>
Unpacks the .traineddata using the provided prefix.</simpara>
</refsect1>
<refsect1 id="_caveats">
<title>CAVEATS</title>
<simpara><emphasis>Prefix</emphasis> refers to the full file prefix, including period (.)</simpara>
</refsect1>
<refsect1 id="_components">
<title>COMPONENTS</title>
<simpara>The components in a Tesseract lang.traineddata file as of
Tesseract 3.02 are briefly described below; For more information on
many of these files, see
<ulink url="https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract">https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract</ulink></simpara>
<variablelist>
<varlistentry>
<term>
lang.config
</term>
<listitem>
<simpara>
(Optional) Language-specific overrides to default config variables.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.unicharset
</term>
<listitem>
<simpara>
(Required) The list of symbols that Tesseract recognizes, with properties.
See unicharset(5).
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.unicharambigs
</term>
<listitem>
<simpara>
(Optional) This file contains information on pairs of recognized symbols
which are often confused. For example, <emphasis>rn</emphasis> and <emphasis>m</emphasis>.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.inttemp
</term>
<listitem>
<simpara>
(Required) Character shape templates for each unichar. Produced by
mftraining(1).
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.pffmtable
</term>
<listitem>
<simpara>
(Required) The number of features expected for each unichar.
Produced by mftraining(1) from <emphasis role="strong">.tr</emphasis> files.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.normproto
</term>
<listitem>
<simpara>
(Required) Character normalization prototypes generated by cntraining(1)
from <emphasis role="strong">.tr</emphasis> files.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.punc-dawg
</term>
<listitem>
<simpara>
(Optional) A dawg made from punctuation patterns found around words.
The "word" part is replaced by a single space.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.word-dawg
</term>
<listitem>
<simpara>
(Optional) A dawg made from dictionary words from the language.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.number-dawg
</term>
<listitem>
<simpara>
(Optional) A dawg made from tokens which originally contained digits.
Each digit is replaced by a space character.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.freq-dawg
</term>
<listitem>
<simpara>
(Optional) A dawg made from the most frequent words which would have
gone into word-dawg.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.fixed-length-dawgs
</term>
<listitem>
<simpara>
(Optional) Several dawgs of different fixed lengths&#8201;&#8212;&#8201;useful for
languages like Chinese.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.cube-unicharset
</term>
<listitem>
<simpara>
(Optional) A unicharset for cube, if cube was trained on a different set
of symbols.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.cube-word-dawg
</term>
<listitem>
<simpara>
(Optional) A word dawg for cube&#8217;s alternate unicharset. Not needed if Cube
was trained with Tesseract&#8217;s unicharset.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.shapetable
</term>
<listitem>
<simpara>
(Optional) When present, a shapetable is an extra layer between the character
classifier and the word recognizer that allows the character classifier to
return a collection of unichar ids and fonts instead of a single unichar-id
and font.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.bigram-dawg
</term>
<listitem>
<simpara>
(Optional) A dawg of word bigrams where the words are separated by a space
and each digit is replaced by a <emphasis>?</emphasis>.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.unambig-dawg
</term>
<listitem>
<simpara>
(Optional) TODO: Describe.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
lang.params-training-model
</term>
<listitem>
<simpara>
(Optional) TODO: Describe.
</simpara>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_history">
<title>HISTORY</title>
<simpara>combine_tessdata(1) first appeared in version 3.00 of Tesseract</simpara>
</refsect1>
<refsect1 id="_see_also">
<title>SEE ALSO</title>
<simpara>tesseract(1), wordlist2dawg(1), cntraining(1), mftraining(1), unicharset(5),
unicharambigs(5)</simpara>
</refsect1>
<refsect1 id="_copying">
<title>COPYING</title>
<simpara>Copyright (C) 2009, Google Inc.
Licensed under the Apache License, Version 2.0</simpara>
</refsect1>
<refsect1 id="_author">
<title>AUTHOR</title>
<simpara>The Tesseract OCR engine was written by Ray Smith and his research groups
at Hewlett Packard (1985-1995) and Google (2006-present).</simpara>
</refsect1>
</refentry>