.\" Man page generated from reStructuredText.
.
.TH OCR4GAMERA 1 "" "" ""
.SH NAME
ocr4gamera \- OCR system using the Gamera framework
.
.nr rst2man-indent-level 0
.
.de1 rstReportMargin
\\$1 \\n[an-margin]
level \\n[rst2man-indent-level]
level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
-
\\n[rst2man-indent0]
\\n[rst2man-indent1]
\\n[rst2man-indent2]
..
.de1 INDENT
.\" .rstReportMargin pre:
. RS \\$1
. nr rst2man-indent\\n[rst2man-indent-level] \\n[an-margin]
. nr rst2man-indent-level +1
.\" .rstReportMargin post:
..
.de UNINDENT
. RE
.\" indent \\n[an-margin]
.\" old: \\n[rst2man-indent\\n[rst2man-indent-level]]
.nr rst2man-indent-level -1
.\" new: \\n[rst2man-indent\\n[rst2man-indent-level]]
.in \\n[rst2man-indent\\n[rst2man-indent-level]]u
..
.SH USAGE
.sp
\fBocr4gamera\fP \-x \fI<traindata>\fP [\fIoptions\fP] \fI<imagefile>\fP
.SH OPTIONS
.INDENT 0.0
.TP
.BI \-v \ <int>\fP,\fB \ \-\-verbosity\fB= <int>
Set verbosity level to \fI<int>\fP\&.
Possible values are
0 (default): silent operation;
1:  information on progress;
>2: segmentation info is written to PNG files with prefix \fBdebug_\fP\&.
.TP
.B \-h\fP,\fB  \-\-help
Display help and exit.
.TP
.B \-\-version
Print version and exit.
.TP
.B \-d\fP,\fB  \-\-deskew
Do a skew correction (recommended).
.UNINDENT
.INDENT 0.0
.TP
.B \-mf <ws>, \-\-median_filter=<ws>
Smooth the input image with a median filter with window size \fI<ws>\fP\&.
Default is \fI<ws>=0\fP, which means no smoothing
.TP
.B \-ds <s>, \-\-despeckle=<s>
Remove all speckle with size \fI<= <s>\fP\&.
Default is \fI<s> = 0\fP, which means no despeckling.
.UNINDENT
.INDENT 0.0
.TP
.B \-f\fP,\fB  \-\-filter
Filter out very large (images) and very small components (noise).
.TP
.B \-a\fP,\fB  \-\-automatic\-group
Autogroup glyphs with classifier.
.TP
.BI \-x \ <file>\fP,\fB \ \-\-xmlfile\fB= <file>
Read training data from \fI<file>\fP\&.
.TP
.BI \-o \ <xml>\fP,\fB \ \-\-output\fB= <xml>
Write recognized text to file \fI<xml>\fP (otherwise it is written to stdout).
.UNINDENT
.INDENT 0.0
.TP
.B \-od <dir>, \-\-output_directory=<dir>
Writes for each input image \fI<img>\fP the recognized text to \fI<dir>/<img>.txt\fP\&.
Note that this option cannot be used in combination with \fB\-o\fP (\-\-outfile).
.UNINDENT
.INDENT 0.0
.TP
.BI \-c \ <csv>\fP,\fB \ \-\-extra_chars_csvfile\fB= <csv>
Read additional class name conversions from file \fI<csv>\fP\&.
\fI<csv>\fP must contain one conversion per line.
.TP
.BI \-R \ <rules>\fP,\fB \ \-\-heuristic_rules\fB= <rules>
Apply heuristic rules \fI<rules>\fP for disambiguation of some chars.
\fI<rules>\fP can be \fBroman\fP (default) or \fBnone\fP (for no rules).
.TP
.B \-D\fP,\fB  \-\-dictionary\-correction
Correct words using a dictionary (requires aspell or ispell).
.TP
.BI \-L \ <lang>\fP,\fB \ \-\-dictionary\-language\fB= <lang>
Use \fI<lang>\fP as language for aspell (when option \fB\-D\fP is set).
.TP
.BI \-e \ <int>\fP,\fB \ \-\-edit\-distance\fB= <int>
Correct words only when edit distance not more than \fI<int>\fP\&.
.TP
.BI \-h\fB o\fP,\fB \ \-\-hocr_out
Writes output as hocr file (only works with the \fB\-o\fP option).
.UNINDENT
.INDENT 0.0
.TP
.B \-hi <hocrfile>, \-\-hocr_in=<hocrfile>
Uses an hocr input file for textline segmentation.
.UNINDENT
.\" Generated by docutils manpage writer.
.