.\" it would be fine ...
.\" Generated automatically from man.1.in by the
.\" configure script.
.\"
.TH ocre 1 "July 21, 2006"
.LO 1
.SH NAME
ocre \- ocr e..... (to fill)
.SH SYNOPSIS
.B ocre
[\fIoptions\fR]\fI nfImage\fR ...
.sp
Options (ephemeral):
.B [\-h]
.BI "[\-d " ndeb ]
.BI "[\-l " language ]
.BI "[\-p " nproc]
.BI "[\-C " char]
.BI "[\-D " ]
.SH DESCRIPTION
.B ocre
recognises characters from the images in the files \fInfImage\fR ... .
The input files must be
in \fBpgm\fR(raw) format (256 gray levels) or
in \fBpbm\fR(raw) format.
At the standard output we get the recognised text.
.SH OPTIONS (Ephemeral)
Some of the present options have been introduced
for help in debugging and
building a font data base
employed to teach the system.
There is no commitment
for supporting them.
.TP
.BI "\-\^d " ndeb
Writes debugging messages at the standard output.
.TP
.B \-\^h
Writes a help message with options.
.TP
.BI "\-\^l " language
Expects \fIlanguage\fR in the document.
Opens a window with specific characters of the language.
Adapts the output to the language.
(for instance ascii 8859-2 if language is 'pol' (polish)).
.TP
.BI "\-\^p " nproc
Process to do with the images.
If 
.I nproc
is 3, shows the segmented characters.
If 
.I nproc
is 5, writes the characters from the image,
each one in a file, under ./arm1 .
.TP
.BI "\-\^C " char
writes a enlarged average version of character \fIchar\fR.
.TP
.B "\-\^D "
uses \fBaspell\fR to fill some words.

.SH FILES
.TP 15
/usr/local/share/ocre/decsWood
Directory with character decision trees.
.TP
/usr/local/share/ocre/decsWood/00index
decision trees index.
.TP
/usr/local/share/ocre/decsWood/arbol_*=??.id3
decision trees.

.SH AUTHOR
Luis Cearra. http://lem.eui.upm.es/luisjc.html

.SH COPYRIGHTS
GPL

.SH ERRORS
Does not recognise well large characters,
very little ones, italic,
two columns, ..., I am working in it .
This is a small model. :-) .