tesseract/doc/unicharset.5.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>UNICHARSET(5)</title>
</refentryinfo>
<refmeta>
<refentrytitle>unicharset</refentrytitle>
<manvolnum>5</manvolnum>
<refmiscinfo class="source">&#160;</refmiscinfo>
<refmiscinfo class="manual">&#160;</refmiscinfo>
</refmeta>
<refnamediv>
    <refname>unicharset</refname>
    <refpurpose>character properties file used by tesseract(1)</refpurpose>
</refnamediv>
<refsect1 id="_description">
<title>DESCRIPTION</title>
<simpara>Tesseract&#8217;s unicharset file contains information on each symbol
(unichar) the Tesseract OCR engine is trained to recognize.</simpara>
<simpara>A unicharset file (i.e. <emphasis>eng.unicharset</emphasis>) is distributed as part of a
Tesseract language pack (i.e. <emphasis>eng.traineddata</emphasis>).  For information on
extracting the unicharset file, see combine_tessdata(1).</simpara>
<simpara>The first line of a unicharset file contains the number of unichars in
the file.  After this line, each subsequent line provides information for
a single unichar.  The first such line contains a placeholder reserved for
the space character.  Each unichar is referred to within Tesseract by its
Unichar ID, which is the line number (minus 1) within the unicharset file.
Therefore, space gets unichar 0.</simpara>
<simpara>Each unichar line in the unicharset file (v2+) may have four space-separated fields:</simpara>
<literallayout class="monospaced">'character' 'properties' 'script' 'id'</literallayout>
<simpara>Starting with Tesseract v3.02, more information may be given for each unichar:</simpara>
<literallayout class="monospaced">'character' 'properties' 'glyph_metrics' 'script' 'other_case' 'direction' 'mirror' 'normed_form'</literallayout>
<simpara>Entries:</simpara>
<variablelist>
<varlistentry>
<term>
<emphasis>character</emphasis>
</term>
<listitem>
<simpara>
The UTF-8 encoded string to be produced for this unichar.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>properties</emphasis>
</term>
<listitem>
<simpara>
An integer mask of character properties, one per bit.
    From least to most significant bit, these are: isalpha, islower, isupper,
    isdigit, ispunctuation.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>glyph_metrics</emphasis>
</term>
<listitem>
<simpara>
Ten comma-separated integers representing various standards
    for where this glyph is to be found within a baseline-normalized coordinate
    system where 128 is normalized to x-height.
</simpara>
<itemizedlist>
<listitem>
<simpara>
min_bottom, max_bottom: the ranges where the bottom of the character can
    be found.
</simpara>
</listitem>
<listitem>
<simpara>
min_top, max_top: the ranges where the top of the character may be found.
</simpara>
</listitem>
<listitem>
<simpara>
min_width, max_width: horizontal width of the character.
</simpara>
</listitem>
<listitem>
<simpara>
min_bearing, max_bearing: how far from the usual start position does the
    leftmost part of the character begin.
</simpara>
</listitem>
<listitem>
<simpara>
min_advance, max_advance: how far from the printer&#8217;s cell left do we
    advance to begin the next character.
</simpara>
</listitem>
</itemizedlist>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>script</emphasis>
</term>
<listitem>
<simpara>
Name of the script (Latin, Common, Greek, Cyrillic, Han, null).
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>other_case</emphasis>
</term>
<listitem>
<simpara>
The Unichar ID of the other case version of this character
    (upper or lower).
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>direction</emphasis>
</term>
<listitem>
<simpara>
The Unicode BiDi direction of this character, as defined by
    ICU&#8217;s enum UCharDirection. (0 = Left to Right, 1 = Right to Left,
    2 = European Number&#8230;)
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>mirror</emphasis>
</term>
<listitem>
<simpara>
The Unichar ID of the BiDirectional mirror of this character.
    For example the mirror of open paren is close paren, but Latin Capital C
    has no mirror, so it remains a Latin Capital C.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>normed_form</emphasis>
</term>
<listitem>
<simpara>
The UTF-8 representation of a "normalized form" of this unichar
    for the purpose of blaming a module for errors given ground truth text.
    For instance, a left or right single quote may normalize to an ASCII quote.
</simpara>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_example_v2">
<title>EXAMPLE (v2)</title>
<literallayout class="monospaced">; 10 Common 46
b 3 Latin 59
W 5 Latin 40
7 8 Common 66
= 0 Common 93</literallayout>
<simpara>";" is a punctuation character. Its properties are thus represented by the
binary number 10000 (10 in hexadecimal).</simpara>
<simpara>"b" is an alphabetic character and a lower case character. Its properties are
thus represented by the binary number 00011 (3 in hexadecimal).</simpara>
<simpara>"W" is an alphabetic character and an upper case character. Its properties are
thus represented by the binary number 00101 (5 in hexadecimal).</simpara>
<simpara>"7" is just a digit. Its properties are thus represented by the binary number
01000 (8 in hexadecimal).</simpara>
<simpara>"=" is not punctuation nor a digit nor an alphabetic character. Its properties
are thus represented by the binary number 00000 (0 in hexadecimal).</simpara>
<simpara>Japanese or Chinese alphabetic character properties are represented by the
binary number 00001 (1 in hexadecimal): they are alphabetic, but neither
upper nor lower case.</simpara>
</refsect1>
<refsect1 id="_example_v3_02">
<title>EXAMPLE (v3.02)</title>
<literallayout class="monospaced">110
NULL 0 NULL 0
N 5 59,68,216,255,87,236,0,27,104,227 Latin 11 0 1 N
Y 5 59,68,216,255,91,205,0,47,91,223 Latin 33 0 2 Y
1 8 59,69,203,255,45,128,0,66,74,173 Common 3 2 3 1
9 8 18,66,203,255,89,156,0,39,104,173 Common 4 2 4 9
a 3 58,65,186,198,85,164,0,26,97,185 Latin 56 0 5 a
. . .</literallayout>
</refsect1>
<refsect1 id="_caveats">
<title>CAVEATS</title>
<simpara>Although the unicharset reader maintains the ability to read unicharsets
of older formats and will assign default values to missing fields,
the accuracy will be degraded.</simpara>
<simpara>Further, most other data files are indexed by the unicharset file,
so changing it without re-generating the others is likely to have dire
consequences.</simpara>
</refsect1>
<refsect1 id="_history">
<title>HISTORY</title>
<simpara>The unicharset format first appeared with Tesseract 2.00, which was the
first version to support languages other than English. The unicharset file
contained only the first two fields, and the "ispunctuation" property was
absent (punctuation was regarded as "0", as "=" is in the above example.</simpara>
</refsect1>
<refsect1 id="_see_also">
<title>SEE ALSO</title>
<simpara>tesseract(1), combine_tessdata(1), unicharset_extractor(1)</simpara>
<simpara><ulink url="https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract">https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract</ulink></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>