cuneiform(1) multi-language OCR system


cuneiform [--dotmatrix] [--fax] [--singlecolumn] [-f format] [-l language] [-o output] input


Cuneiform is an OCR system. In addition to text recognition it also does layout analysis and text format recognition. Cuneiform supports several languages.


Use recognition mode optimized for text printed with a dot matrix printer.
Use recognition mode optimized for text that has been faxed.
Disable page layout analysis and assumes that the image consists of only one column of text.
-f format
Select output format. The following formats are available: html (HTML format), hocr (hOCR HTML format), native (native Cuneiform 2000), rtf (RTF format), smarttext (plain text with TeX paragraphs), text (plain text). The default is plain text.
-l language
By default Cuneiform recognizes English text. To change the language use the command line switch -l followed by a language code (typically an ISO 639-2 three-letter code). The following languages are supported:
bul Bulgarian
cze Czech
dan Danish
dut Dutch
eng English
est Estonian
fra French
ger German
hrv Croatian
hun Hungarian
ita Italian
lav Latvian
lit Lithuanian
pol Polish
por Portugese
rum Romanian
rus Russian
ruseng mixed Russian/English
slv Slovenian
spa Spanish
srp Serbian
swe Swedish
tur Turkish
ukr Ukrainian
-o output
If you do not define an output file with the -o switch, Cuneiform writes the result to a file 'cuneiform-out.format'. The file extension depends on your output format.


Cuneiform can process any single-page image that GraphicsMagick knows how to open. Please consult the gm(1) manual page for the comprehensive list of supported image formats.


More information about cuneiform can be found at <>.


cuneiform was written by Cognitive Technologies and Jussi Pakkanen <[email protected]>.

This manual page was written by Daniel Baumann <[email protected]>, for the Debian project (but may be used by others).