Re: [orca-list] Getting and reading scanned images with Orca

["John G. Heim" <jheim math wisc edu>]

You use sane to operate your scanner. Here's a link to a script that I 
use to scan and read a document on my  system. It calls  the  scanimage 
command from the sane package, then it uses tesseract to do OCR, then it 
mores the text file that is created so your screen reader will speak it. 
If you put the script in /usr/local/bin, all you have to do is type 
readdoc and wait for it to start talking.

http://www.johnheim.net/~john/Downloads/readdoc

#!/bin/bash

MYIMAGE=/tmp/page.tif
MYTEXT=/tmp/page
MYRES=300

if test ! -z $1; then
        MYRES=$1
fi

scanimage --format=tiff --mode Lineart --resolution $MYRES > $MYIMAGE
tesseract $MYIMAGE $MYTEXT
more "$MYTEXT.txt"





On 12/10/13 10:46, luciano de souza wrote:
> Yes, it seemed to be very well.
>
> Can Tesseract be used also to call the scanner to capture the pages or
> is it only a very nice OCR?
>
> 2013/12/10, luciano de souza <luchyanus gmail com>:
> > Hello all,
> >
> > With some urgency, I need to read a scanned document. An image was
> > sent by e-mail. As a result, I need a OCR processor.
> >
> > I heard that free OCR software in Linux is not so acurate than it is
> > in Windows. This is not the problem. Today, I need to read the test
> > and I have only Linux in my machine.
> >
> > So I ask: which packages can I install in order to process the image
> > to extract the test? Is it something easy to use?
> >
> > --
> > Luciano de Souza

--
---
John G. Heim, 608-263-4189, jheim math wisc edu