tesseract/doc/tesseract.1.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>TESSERACT(1)</title>
</refentryinfo>
<refmeta>
<refentrytitle>tesseract</refentrytitle>
<manvolnum>1</manvolnum>
<refmiscinfo class="source">&#160;</refmiscinfo>
<refmiscinfo class="manual">&#160;</refmiscinfo>
</refmeta>
<refnamediv>
    <refname>tesseract</refname>
    <refpurpose>command-line OCR engine</refpurpose>
</refnamediv>
<refsynopsisdiv id="_synopsis">
<simpara><emphasis role="strong">tesseract</emphasis> <emphasis>imagename</emphasis>|<emphasis>stdin</emphasis> <emphasis>outputbase</emphasis>|<emphasis>stdout</emphasis> [options&#8230;] [configfile&#8230;]</simpara>
</refsynopsisdiv>
<refsect1 id="_description">
<title>DESCRIPTION</title>
<simpara>tesseract(1) is a commercial quality OCR engine originally developed at HP
between 1985 and 1995. In 1995, this engine was among the top 3 evaluated by
UNLV. It was open-sourced by HP and UNLV in 2005, and has been developed
at Google since then.</simpara>
</refsect1>
<refsect1 id="_in_out_arguments">
<title>IN/OUT ARGUMENTS</title>
<variablelist>
<varlistentry>
<term>
<emphasis>imagename</emphasis>
</term>
<listitem>
<simpara>
        The name of the input image.  Most image file formats (anything
        readable by Leptonica) are supported.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>stdin</emphasis>
</term>
<listitem>
<simpara>
        Instruction to read data from standard input
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>outputbase</emphasis>
</term>
<listitem>
<simpara>
        The basename of the output file (to which the appropriate extension
        will be appended).  By default the output will be named <emphasis>outbase.txt</emphasis>.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>stdout</emphasis>
</term>
<listitem>
<simpara>
        Instruction to sent output data to standard output
</simpara>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_options">
<title>OPTIONS</title>
<variablelist>
<varlistentry>
<term>
<emphasis>--tessdata-dir /path</emphasis>
</term>
<listitem>
<simpara>
        Specify the location of tessdata path
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>--user-words /path/to/file</emphasis>
</term>
<listitem>
<simpara>
        Specify the location of user words file
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>--user-patterns /path/to/file specify</emphasis>
</term>
<listitem>
<simpara>
        The location of user patterns file
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>-c configvar=value</emphasis>
</term>
<listitem>
<simpara>
        Set value for control parameter. Multiple -c arguments are allowed.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>-l lang</emphasis>
</term>
<listitem>
<simpara>
        The language to use. If none is specified, English is assumed.
        Multiple languages may be specified, separated by plus characters.
        Tesseract uses 3-character ISO 639-2 language codes. (See LANGUAGES)
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>--psm N</emphasis>
</term>
<listitem>
<simpara>
        Set Tesseract to only run a subset of layout analysis and assume
        a certain form of image. The options for <emphasis role="strong">N</emphasis> are:
</simpara>
<literallayout class="monospaced">0 = Orientation and script detection (OSD) only.
1 = Automatic page segmentation with OSD.
2 = Automatic page segmentation, but no OSD, or OCR.
3 = Fully automatic page segmentation, but no OSD. (Default)
4 = Assume a single column of text of variable sizes.
5 = Assume a single uniform block of vertically aligned text.
6 = Assume a single uniform block of text.
7 = Treat the image as a single text line.
8 = Treat the image as a single word.
9 = Treat the image as a single word in a circle.
10 = Treat the image as a single character.</literallayout>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>configfile</emphasis>
</term>
<listitem>
<simpara>
        The name of a config to use. A config is a plaintext file which
        contains a list of variables and their values, one per line, with a
        space separating variable from value.  Interesting config files
        include:<?asciidoc-br?>
</simpara>
<itemizedlist>
<listitem>
<simpara>
hocr - Output in hOCR format instead of as a text file.
</simpara>
</listitem>
<listitem>
<simpara>
pdf  - Output in pdf instead of a text file.
</simpara>
</listitem>
</itemizedlist>
</listitem>
</varlistentry>
</variablelist>
<simpara><emphasis role="strong">Nota Bene:</emphasis>   The options <emphasis>-l lang</emphasis> and <emphasis>--psm N</emphasis> must occur
before any <emphasis>configfile</emphasis>.</simpara>
</refsect1>
<refsect1 id="_single_options">
<title>SINGLE OPTIONS</title>
<variablelist>
<varlistentry>
<term>
<emphasis>-v</emphasis>
</term>
<listitem>
<simpara>
        Returns the current version of the tesseract(1) executable.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>--list-langs</emphasis>
</term>
<listitem>
<simpara>
        list available languages for tesseract engine. Can be used with --tessdata-dir.
</simpara>
</listitem>
</varlistentry>
<varlistentry>
<term>
<emphasis>--print-parameters</emphasis>
</term>
<listitem>
<simpara>
        print tesseract parameters to the stdout.
</simpara>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1 id="_languages">
<title>LANGUAGES</title>
<simpara>There are currently language packs available for the following languages
(in <ulink url="https://github.com/tesseract-ocr/tessdata">https://github.com/tesseract-ocr/tessdata</ulink>):</simpara>
<simpara><emphasis role="strong">afr</emphasis> (Afrikaans)
<emphasis role="strong">amh</emphasis> (Amharic)
<emphasis role="strong">ara</emphasis> (Arabic)
<emphasis role="strong">asm</emphasis> (Assamese)
<emphasis role="strong">aze</emphasis> (Azerbaijani)
<emphasis role="strong">aze_cyrl</emphasis> (Azerbaijani - Cyrilic)
<emphasis role="strong">bel</emphasis> (Belarusian)
<emphasis role="strong">ben</emphasis> (Bengali)
<emphasis role="strong">bod</emphasis> (Tibetan)
<emphasis role="strong">bos</emphasis> (Bosnian)
<emphasis role="strong">bul</emphasis> (Bulgarian)
<emphasis role="strong">cat</emphasis> (Catalan; Valencian)
<emphasis role="strong">ceb</emphasis> (Cebuano)
<emphasis role="strong">ces</emphasis> (Czech)
<emphasis role="strong">chi_sim</emphasis> (Chinese - Simplified)
<emphasis role="strong">chi_tra</emphasis> (Chinese - Traditional)
<emphasis role="strong">chr</emphasis> (Cherokee)
<emphasis role="strong">cym</emphasis> (Welsh)
<emphasis role="strong">dan</emphasis> (Danish)
<emphasis role="strong">dan_frak</emphasis> (Danish - Fraktur)
<emphasis role="strong">deu</emphasis> (German)
<emphasis role="strong">deu_frak</emphasis> (German - Fraktur)
<emphasis role="strong">dzo</emphasis> (Dzongkha)
<emphasis role="strong">ell</emphasis> (Greek, Modern (1453-))
<emphasis role="strong">eng</emphasis> (English)
<emphasis role="strong">enm</emphasis> (English, Middle (1100-1500))
<emphasis role="strong">epo</emphasis> (Esperanto)
<emphasis role="strong">equ</emphasis> (Math / equation detection module)
<emphasis role="strong">est</emphasis> (Estonian)
<emphasis role="strong">eus</emphasis> (Basque)
<emphasis role="strong">fas</emphasis> (Persian)
<emphasis role="strong">fin</emphasis> (Finnish)
<emphasis role="strong">fra</emphasis> (French)
<emphasis role="strong">frk</emphasis> (Frankish)
<emphasis role="strong">frm</emphasis> (French, Middle (ca.1400-1600))
<emphasis role="strong">gle</emphasis> (Irish)
<emphasis role="strong">glg</emphasis> (Galician)
<emphasis role="strong">grc</emphasis> (Greek, Ancient (to 1453))
<emphasis role="strong">guj</emphasis> (Gujarati)
<emphasis role="strong">hat</emphasis> (Haitian; Haitian Creole)
<emphasis role="strong">heb</emphasis> (Hebrew)
<emphasis role="strong">hin</emphasis> (Hindi)
<emphasis role="strong">hrv</emphasis> (Croatian)
<emphasis role="strong">hun</emphasis> (Hungarian)
<emphasis role="strong">iku</emphasis> (Inuktitut)
<emphasis role="strong">ind</emphasis> (Indonesian)
<emphasis role="strong">isl</emphasis> (Icelandic)
<emphasis role="strong">ita</emphasis> (Italian)
<emphasis role="strong">ita_old</emphasis> (Italian - Old)
<emphasis role="strong">jav</emphasis> (Javanese)
<emphasis role="strong">jpn</emphasis> (Japanese)
<emphasis role="strong">kan</emphasis> (Kannada)
<emphasis role="strong">kat</emphasis> (Georgian)
<emphasis role="strong">kat_old</emphasis> (Georgian - Old)
<emphasis role="strong">kaz</emphasis> (Kazakh)
<emphasis role="strong">khm</emphasis> (Central Khmer)
<emphasis role="strong">kir</emphasis> (Kirghiz; Kyrgyz)
<emphasis role="strong">kor</emphasis> (Korean)
<emphasis role="strong">kur</emphasis> (Kurdish)
<emphasis role="strong">lao</emphasis> (Lao)
<emphasis role="strong">lat</emphasis> (Latin)
<emphasis role="strong">lav</emphasis> (Latvian)
<emphasis role="strong">lit</emphasis> (Lithuanian)
<emphasis role="strong">mal</emphasis> (Malayalam)
<emphasis role="strong">mar</emphasis> (Marathi)
<emphasis role="strong">mkd</emphasis> (Macedonian)
<emphasis role="strong">mlt</emphasis> (Maltese)
<emphasis role="strong">msa</emphasis> (Malay)
<emphasis role="strong">mya</emphasis> (Burmese)
<emphasis role="strong">nep</emphasis> (Nepali)
<emphasis role="strong">nld</emphasis> (Dutch; Flemish)
<emphasis role="strong">nor</emphasis> (Norwegian)
<emphasis role="strong">ori</emphasis> (Oriya)
<emphasis role="strong">osd</emphasis> (Orientation and script detection module)
<emphasis role="strong">pan</emphasis> (Panjabi; Punjabi)
<emphasis role="strong">pol</emphasis> (Polish)
<emphasis role="strong">por</emphasis> (Portuguese)
<emphasis role="strong">pus</emphasis> (Pushto; Pashto)
<emphasis role="strong">ron</emphasis> (Romanian; Moldavian; Moldovan)
<emphasis role="strong">rus</emphasis> (Russian)
<emphasis role="strong">san</emphasis> (Sanskrit)
<emphasis role="strong">sin</emphasis> (Sinhala; Sinhalese)
<emphasis role="strong">slk</emphasis> (Slovak)
<emphasis role="strong">slk_frak</emphasis> (Slovak - Fraktur)
<emphasis role="strong">slv</emphasis> (Slovenian)
<emphasis role="strong">spa</emphasis> (Spanish; Castilian)
<emphasis role="strong">spa_old</emphasis> (Spanish; Castilian - Old)
<emphasis role="strong">sqi</emphasis> (Albanian)
<emphasis role="strong">srp</emphasis> (Serbian)
<emphasis role="strong">srp_latn</emphasis> (Serbian - Latin)
<emphasis role="strong">swa</emphasis> (Swahili)
<emphasis role="strong">swe</emphasis> (Swedish)
<emphasis role="strong">syr</emphasis> (Syriac)
<emphasis role="strong">tam</emphasis> (Tamil)
<emphasis role="strong">tel</emphasis> (Telugu)
<emphasis role="strong">tgk</emphasis> (Tajik)
<emphasis role="strong">tgl</emphasis> (Tagalog)
<emphasis role="strong">tha</emphasis> (Thai)
<emphasis role="strong">tir</emphasis> (Tigrinya)
<emphasis role="strong">tur</emphasis> (Turkish)
<emphasis role="strong">uig</emphasis> (Uighur; Uyghur)
<emphasis role="strong">ukr</emphasis> (Ukrainian)
<emphasis role="strong">urd</emphasis> (Urdu)
<emphasis role="strong">uzb</emphasis> (Uzbek)
<emphasis role="strong">uzb_cyrl</emphasis> (Uzbek - Cyrilic)
<emphasis role="strong">vie</emphasis> (Vietnamese)
<emphasis role="strong">yid</emphasis> (Yiddish)</simpara>
<simpara>To use a non-standard language pack named <emphasis role="strong">foo.traineddata</emphasis>, set the
<emphasis role="strong">TESSDATA_PREFIX</emphasis> environment variable so the file can be found at
<emphasis role="strong">TESSDATA_PREFIX</emphasis>/tessdata/<emphasis role="strong">foo</emphasis>.traineddata and give Tesseract the
argument <emphasis>-l foo</emphasis>.</simpara>
</refsect1>
<refsect1 id="_config_files_and_augmenting_with_user_data">
<title>CONFIG FILES AND AUGMENTING WITH USER DATA</title>
<simpara>Tesseract config files consist of lines with variable-value pairs (space
separated).  The variables are documented as flags in the source code like
the following one in tesseractclass.h:</simpara>
<simpara>STRING_VAR_H(tessedit_char_blacklist, "",
             "Blacklist of chars not to recognize");</simpara>
<simpara>These variables may enable or disable various features of the engine, and
may cause it to load (or not load) various data.  For instance, let&#8217;s suppose
you want to OCR in English, but suppress the normal dictionary and load an
alternative word list and an alternative list of patterns&#8201;&#8212;&#8201;these two files
are the most commonly used extra data files.</simpara>
<simpara>If your language pack is in /path/to/eng.traineddata  and the hocr config
is in /path/to/configs/hocr then create three new files:</simpara>
<simpara>/path/to/eng.user-words:</simpara>
<blockquote>
<literallayout>the
quick
brown
fox
jumped</literallayout>
</blockquote>
<simpara>/path/to/eng.user-patterns:</simpara>
<blockquote>
<literallayout>1-\d\d\d-GOOG-411
www.\n\\\*.com</literallayout>
</blockquote>
<simpara>/path/to/configs/bazaar:</simpara>
<blockquote>
<literallayout>load_system_dawg     F
load_freq_dawg       F
user_words_suffix    user-words
user_patterns_suffix user-patterns</literallayout>
</blockquote>
<simpara>Now, if you pass the word <emphasis>bazaar</emphasis> as a trailing command line parameter
to Tesseract, Tesseract will not bother loading the system dictionary nor
the dictionary of frequent words and will load and use the eng.user-words
and eng.user-patterns files you provided.  The former is a simple word list,
one per line.  The format of the latter is documented in dict/trie.h
on read_pattern_list().</simpara>
</refsect1>
<refsect1 id="_history">
<title>HISTORY</title>
<simpara>The engine was developed at Hewlett Packard Laboratories Bristol and at
Hewlett Packard Co, Greeley Colorado between 1985 and 1994, with some more
changes made in 1996 to port to Windows, and some C++izing in 1998. A
lot of the code was written in C, and then some more was written in C++.
The C\++ code makes heavy use of a list system using macros. This predates
stl, was portable before stl, and is more efficient than stl lists, but has
the big negative that if you do get a segmentation violation, it is hard to
debug.</simpara>
<simpara>Version 2.00 brought Unicode (UTF-8) support, six languages, and the ability
to train Tesseract.</simpara>
<simpara>Tesseract was included in UNLV&#8217;s Fourth Annual Test of OCR Accuracy.
See <ulink url="https://github.com/tesseract-ocr/docs/blob/master/AT-1995.pdf">https://github.com/tesseract-ocr/docs/blob/master/AT-1995.pdf</ulink>. With Tesseract 2.00,
scripts are now included to allow anyone to reproduce some of these tests.
See <ulink url="https://github.com/tesseract-ocr/tesseract/wiki/TestingTesseract">https://github.com/tesseract-ocr/tesseract/wiki/TestingTesseract</ulink> for more
details.</simpara>
<simpara>Tesseract 3.00 adds a number of new languages, including Chinese, Japanese,
and Korean. It also introduces a new, single-file based system of managing
language data.</simpara>
<simpara>Tesseract 3.02 adds BiDirectional text support, the ability to recognize
multiple languages in a single image, and improved layout analysis.</simpara>
<simpara>For further details, see the file ReleaseNotes included with the distribution.</simpara>
</refsect1>
<refsect1 id="_resources">
<title>RESOURCES</title>
<simpara>Main web site: <ulink url="https://github.com/tesseract-ocr">https://github.com/tesseract-ocr</ulink><?asciidoc-br?>
Information on training: <ulink url="https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract">https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract</ulink></simpara>
</refsect1>
<refsect1 id="_see_also">
<title>SEE ALSO</title>
<simpara>ambiguous_words(1), cntraining(1), combine_tessdata(1), dawg2wordlist(1),
shape_training(1), mftraining(1), unicharambigs(5), unicharset(5),
unicharset_extractor(1), wordlist2dawg(1)</simpara>
</refsect1>
<refsect1 id="_author">
<title>AUTHOR</title>
<simpara>Tesseract development was led at Hewlett-Packard and Google by Ray Smith.
The development team has included:</simpara>
<simpara>Ahmad Abdulkader, Chris Newton, Dan Johnson, Dar-Shyang Lee, David Eger,
Eric Wiseblatt, Faisal Shafait, Hiroshi Takenaka, Joe Liu, Joern Wanke,
Mark Seaman, Mickey Namiki, Nicholas Beato, Oded Fuhrmann, Phil Cheatle,
Pingping Xiu, Pong Eksombatchai (Chantat), Ranjith Unnikrishnan, Raquel
Romano, Ray Smith, Rika Antonova, Robert Moss, Samuel Charron, Sheelagh
Lloyd, Shobhit Saxena, and Thomas Kielbus.</simpara>
</refsect1>
<refsect1 id="_copying">
<title>COPYING</title>
<simpara>Licensed under the Apache License, Version 2.0</simpara>
</refsect1>
</refentry>