Nabble - Gnu - Ocrad

Re: Library version of OCRAD

2009-11-25T10:08:58Z

Igor Filippov wrote:
> The author of OSRA not only finds the idea useful, I have asked for such
> a library before on this very list. So I am very happy to help in any
> way possible to bring this idea to reality!

Then I'll study OSRA's source to see the best way of developing an
interface. Please, send me any idea you have about this. I think I
didn't know about OSRA before, but I find it a very interesting and
useful package.

Best regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Library version of OCRAD

2009-11-25T09:32:58Z

The author of OSRA not only finds the idea useful, I have asked for such
a library before on this very list. So I am very happy to help in any
way possible to bring this idea to reality!

Regards,
Igor

On Wed, 2009-11-25 at 18:14 +0100, Antonio Diaz Diaz wrote:

> Konstantin Tokarev wrote:
> > I think it would be good idea to release OCRAD library. For example, OSRA (Optical Structure Recognition Applilcation, http://cactus.nci.nih.gov/osra/#7 ) actually uses OCRAD source without main() function
>
> Thanks for the link. I'll ask the authors of OSRA if they find the idea
> useful and what would be the best interface.
>
>
> Best regards,
> Antonio.
>
>
> _______________________________________________
> Bug-ocrad mailing list
> Bug-ocrad@...
> http://lists.gnu.org/mailman/listinfo/bug-ocrad

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Library version of OCRAD

2009-11-25T09:14:00Z

Konstantin Tokarev wrote:
> I think it would be good idea to release OCRAD library. For example, OSRA (Optical Structure Recognition Applilcation, http://cactus.nci.nih.gov/osra/#7 ) actually uses OCRAD source without main() function

Thanks for the link. I'll ask the authors of OSRA if they find the idea
useful and what would be the best interface.

Best regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Library version of OCRAD

2009-11-24T07:52:12Z

I think it would be good idea to release OCRAD library. For example, OSRA (Optical Structure Recognition Applilcation, http://cactus.nci.nih.gov/osra/#7 ) actually uses OCRAD source without main() function
--
Regards,
Konstantin

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Please could you help me installing OCRAD ?

2009-10-26T07:34:50Z

Hello,

Antoine Morcos [Gamekyo.com] wrote:
> I'm traying to install Ocrad on my Linux server. but I get an error message
> when a launch Make :
>
> rational.cc: In method `int Rational::parse (const char *)':
> rational.cc:151: parse error before `__ctype_b'

It seems your cctype header defines std::isspace and std::isdigit as
macros. What version of g++ and glibc are you using? IIRC, ocrad needs
at least g++ 3.0 and glibc 2.1.

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Please could you help me installing OCRAD ?

2009-10-25T16:43:18Z

Hi,

I'm traying to install Ocrad on my Linux server. but I get an error message
when a launch Make :

I get *make: *** [rational.o] Erreur 1*

[root@ks355185 ocrad]# cd ocrad-0.18
[root@ks355185 ocrad-0.18]# ./configure
creating config.status
creating Makefile
VPATH = .
prefix = /usr/local
exec_prefix = $(prefix)
bindir = $(exec_prefix)/bin
datadir = $(prefix)/share
infodir = $(datadir)/info
mandir = $(datadir)/man
sysconfdir = $(prefix)/etc
CC = gcc
CXX = g++
CPPFLAGS =
CFLAGS = -Wall -W -O2
CXXFLAGS = -Wall -W -O2
LDFLAGS =
OK. Now you can run make.
[root@ks355185 ocrad-0.18]# make
g++ -Wall -W -O2 -c -o arg_parser.o arg_parser.cc
g++ -Wall -W -O2 -c -o common.o common.cc
g++ -Wall -W -O2 -c -o mask.o mask.cc
g++ -Wall -W -O2 -c -o rational.o rational.cc
rational.cc: In method `int Rational::parse (const char *)':
rational.cc:151: parse error before `__ctype_b'
rational.cc:154: parse error before `__ctype_b'
rational.cc:156: parse error before `__ctype_b'
rational.cc:160: confused by earlier errors, bailing out
*make: *** [rational.o] Erreur 1*

Do you have an idea what it is and how I can pass this error ?

Thank you very much for your help.

Antoine
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Training for new language

2009-09-09T02:42:03Z

Hi,

First of all I would to apologize from you because this message isn't
real bug, but I forced to post this message because I couldn't find any
support resource except this mailing list.

I just want to ask you a tiny question:

Is there any ability to train ocrad for new language? if yes, what's the
limitation for characters number?

Thanks.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

OCR beginer

2009-08-17T04:21:46Z

Hello,

my name is Vaclav Kocian and on september I will start my Phd program on
Ostrava university (http://prf.osu.eu/kip/). Theme of my disertation is
about artifical intelligence in patern recognition, focused in ancient text
digitalization. In my opinion cooperation on some OCR project can be usefull
for me as for the project itself (as I belive). So I'm interesting in
posibility of join Ocrad team. I also hope, that you could be so kind and
let me know about some valuable sources of literature about text
recognition.

With regards. V.K.
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Source Code Documentation

2009-08-17T04:19:42Z

Sir,
I want to use your excellent program for recognizing devnagari characters (Sanskrit), suitably modifying it. I've gone through all the classes the program consists, however it'll be highly beneficial if official developer's documentations is availed. If such documentation or connotations are available. Please let me know.

Thanking you.

Yogesh Kumbhar

Looking for local information? Find it on Yahoo! Local http://in.local.yahoo.com/
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

ocrad usage

2009-07-25T13:15:10Z

Dear bugtrackers and developers,

I am trying to use the gnu ocr OCRAD to extract text from scanned
documents. Reviews of the software deem it to be "reasonably good" and
to "produce fairly accurate results". Unfortunately, when I use OCRAD
to parse images, I do not even get any barely usable results. The
output of OCRAD looks more like a dumped gpg encrypted file then a
document - I'm serious, not even remotely readable.

I have tried everything I could think of. I printed the "quick brown
fox jumps over the lazy dog" in Arial and New Times Roman, size ranging
from 9 to 16 on an A4 paper and scanned it in color, grayscale and
black-and-white, with 72, 300, 750 and 1200 dpi. The 12 scanned images
each got saved as pbm and ppm. that makes 24 files and not even one was
processed by ocrad to produce remotely readable results. The best
approximation was "qa\;c_br0mfox ipmpsO wer the |psYdOq", by processing
the 750 dpi grayscale pbm...

Obviously, I'm doing something wrong here, but I don't know what. I am
using kooka to scan the images from a HP Deskjet 4620F. Ocrad is
version 0.17, running on SUSE 11.1 .

If you could hint me to what I am doing wrong here, please do...

thanks for your help in advance

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Binary not found message

2009-07-15T06:28:29Z

jchemconsulting@... wrote:
> I am using Kooka for scanning, but when I try to OCR the program indicates:
>
> Using OCRAD binary: Not found

The problem here is either ocrad is not installed on your system or the
path to the ocrad executable (binary) is not correctly set in kooka.

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Binary not found message

2009-07-14T18:16:47Z

Hello.
I am relatively new to Linux, but have been, for me, surprisingly
successful, with most of the Mepis 8 program. I am using Kooka for
scanning, but when I try to OCR the program indicates:

Using OCRAD binary: Not found

it then shows rolling wheels indefinitely. I reloaded the program
again using Synaptic, but it still did not work. I was careful to scan
to.pbm.

Any suggestions would be most welcome.
John

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: submit corrected output to OCRAD?

2009-07-13T07:19:56Z

Joe Corneli wrote:
> to be able to submit corrected output back to
> OCRAD, so in the future it will have a better
> sense of what characters it was supposed to
> recognize!
>
> What do you think about the possibilities here?

Given that ocrad is not trainable, the possibilities are currently
scarce. But you can send a sample text to my private address and I'll
try to make the next release of ocrad recognize it correctly.

Best regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

submit corrected output to OCRAD?

2009-07-12T13:29:57Z

Ocrad seems to do surprisingly well recognizing
text that was originally generated on my manual
typewriter -- but not perfectly. It doesn't seem
like there's a way to do it at present, but a feature
that would be nice would be --

to be able to submit corrected output back to
OCRAD, so in the future it will have a better
sense of what characters it was supposed to
recognize!

What do you think about the possibilities here?

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Convert brush script image data

2009-06-07T10:05:26Z

Hello,

Anoop Valli wrote:
> How can i convert the image data being in brush script i hope you will
> send me the instruction for this problem my os is debian

Ocrad can't recognize casual scripts like brush script. Sorry.

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Convert brush script image data

2009-06-07T03:42:46Z

How can i convert the image data being in brush script i hope you will
send me the instruction for this problem my os is debian

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Version 0.18 of GNU Ocrad released

2009-05-10T08:47:39Z

I am pleased to announce the release of GNU Ocrad 0.18.

Ocrad is the GNU OCR (Optical Character Recognition) program.

The homepage is at http://www.gnu.org/software/ocrad/ocrad.html

The sources can be dowloaded from http://ftp.gnu.org/gnu/ocrad/ or from
your favorite GNU mirror.

This version is also available in lzip format. If your distro doesn't
yet distribute the lzip program, you can download it from
http://www.nongnu.org/lzip/lzip.html

The md5sums are:
5d2d39e218d7f219e53c08ce07556d66 ocrad-0.18.tar.gz
ba812571391857188b454c436418ec1f ocrad-0.18.tar.lz

This release is also GPG signed. You can download the signature by
appending ".sig" to the URL.

Changes in version 0.18:

* The layout analyser is now able to separate blocks of text of
arbitrary shape as long as they are clearly delimited by white space.

* The new option "--quiet" has been added.

* The "--layout" option no more accepts an argument.

* The "--crop" option now accepts negative coordinates, interpreted
as relative to the right-bottom corner of the image.

* This version recognizes the letter 'a' with ring above.

* Recognition of files with a single big character has been fixed.

* A bug that didn't write maxval when saving pgm or ppm files has
been fixed.

* A problem that prevented compilation with GCC 4.3.0 has been fixed.

* "make install-info" should now work on Debian and OS X.

* The man page is now installed by default.

* Arg_parser has been updated to 1.2.

Please send bug reports and suggestions to bug-ocrad@...

Regards,
Antonio Diaz, GNU Ocrad author and maintainer.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Version 0.18-rc1 of GNU Ocrad released

2009-04-27T16:03:51Z

Version 0.18-rc1 of GNU Ocrad is ready for testing here
http://download.savannah.gnu.org/releases/ocrad/ocrad-0.18-rc1.tar.gz
and here (in lzip format)
http://download.savannah.gnu.org/releases/ocrad/ocrad-0.18-rc1.tar.lz

The md5sums are:
8b91103824c920e602fe0cc9ecd7a70c ocrad-0.18-rc1.tar.gz
03e0ca53c54b43e302fdca199d8563b1 ocrad-0.18-rc1.tar.lz

Please, test it and report any bugs you find.

You can download lzip from http://www.nongnu.org/lzip/lzip.html

Changes in this version:

* Recognition of files with a single big character has been fixed.

* "make install-info" should now work on Debian and OS X.

* The man page is now installed by default.

Regards,
Antonio Diaz, GNU Ocrad author and maintainer.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

MANUAL

2008-12-22T02:37:02Z

Hello, World!

How can i get manuals of OCRAD TO OCRAD by OCRAD?

i research OCR /

THank You for OPEN SOURCE - THANK YOU VERY MUCH

Student Kapakly I.V.
Physics and Mathematics Facility
Co-head IT
pedagogical and psychological sp 504
Government University by T.G.Shevchencko
Tiraspol, Transniestria
MD3300

p/S/
Thanks.

--
Sambist
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Version 0.18-pre4 of GNU Ocrad released

2008-10-11T05:46:25Z

Version 0.18-pre4 of GNU Ocrad is ready for testing here
http://download.savannah.gnu.org/releases/ocrad/ocrad-0.18-pre4.tar.bz2
or here (in lzip format)
http://download.savannah.gnu.org/releases/ocrad/ocrad-0.18-pre4.tar.lz

You can download lzip from http://freshmeat.net/projects/lzip/

Please, test it and report any bugs you find.

Changes in this version:

* The new option "--quiet" has been added.

* Verbosity control of messages has been modified.

Regards,
Antonio Diaz, GNU Ocrad author and maintainer.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Version 0.18-pre3 of GNU Ocrad released

2008-08-21T08:13:22Z

Version 0.18-pre3 of GNU Ocrad is ready for testing here
http://es.geocities.com/ant_diaz2001/ocrad-0.18-pre3.tar.gz

Please, test it and report any bugs you find.

Changes in this version:

* This version recognizes the letter 'a' with ring above.

* Some missing headers that prevented compilation with GCC 4.3.0 have
been included.

Regards,
Antonio Diaz, GNU Ocrad author and maintainer.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: compilation fails on Fedora Core 9

2008-08-12T07:12:20Z

Igor Filippov [Contr] wrote:
> Could it be a problem with gcc 4.3.0?

Yes, it is.

I am making a new release in a couple days fixing it.

Thanks,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

compilation fails on Fedora Core 9

2008-08-12T04:39:46Z

Could it be a problem with gcc 4.3.0?
Here is the log file:

OK. Now you can run make.
[igor@xaero ocrad-0.17]$ make
g++ -Wall -W -O2 -c -o arg_parser.o arg_parser.cc
g++ -Wall -W -O2 -c -o common.o common.cc
common.cc: In member function ‘bool Charset::enable(const char*)’:
common.cc:84: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘void Charset::show_error(const char*,
const char*) const’:
common.cc:106: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘bool Filter::set(const char*)’:
common.cc:118: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘void Filter::show_error(const char*,
const char*) const’:
common.cc:126: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘bool Transformation::set(const char*)’:
common.cc:138: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘void Transformation::show_error(const
char*, const char*) const’:
common.cc:146: error: ‘strcmp’ is not a member of ‘std’
common.cc: In member function ‘bool Control::set_format(const char*)’:
common.cc:157: error: ‘strcmp’ is not a member of ‘std’
common.cc:158: error: ‘strcmp’ is not a member of ‘std’
make: *** [common.o] Error 1

--
Igor Filippov [Contr] <igorf@...>

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: ubuntu 8.04 installation error.

2008-08-06T02:25:23Z

On-line buying account wrote:
> I'm trying to install OCRAD on Ubuntu Hardy. When the install script
> (after make install) gets to:
>
> install-info /usr/local/share/info/ocrad.info /usr/local/share/info/dir
>
> I get this error:
>
> too many arguments

This question is already answered here
http://lists.gnu.org/archive/html/bug-ocrad/2005-11/msg00001.html

But just in case, what version of install-info have you installed?

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

ubuntu 8.04 installation error.

2008-08-05T12:15:24Z

Hi,

I'm trying to install OCRAD on Ubuntu Hardy. When the install script
(after make install) gets to:

install-info /usr/local/share/info/ocrad.info /usr/local/share/info/dir

I get this error:

too many arguments

Help!

Chris Morris

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

A couple of "unrecognizable" digits

2008-07-20T12:04:18Z

Hello,

Here is a couple of images that ocrad 0.17 can't recognize properly (empty
output):
http://cmpd2.phys.msu.ru/~arseniy/twoo.ppm
http://cmpd2.phys.msu.ru/~arseniy/foour.ppm

Maybe it'll help improving recognition quality.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: Problems with "." (Point) and the Font "Tahoma"

2008-06-19T06:37:21Z

Klemens Loschy wrote:
> ocrad always detects a "," (comma) when a "." (point) is shown in the
> picture (font is Tahoma = std. Windows font). The point is printed with
> two pixels one upon the other so i think this might cause this problem.
> Is there a way to read the point as point instead of reading a comma?

I can't be sure without seeing a sample image, but if with "two pixels
one upon the other" you mean "the height of the point is twice its
width", then ocrad will always recognize it as a comma.

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Problems with "." (Point) and the Font "Tahoma"

2008-06-19T03:03:37Z

Hi,

ocrad always detects a "," (comma) when a "." (point) is shown in the
picture (font is Tahoma = std. Windows font). The point is printed with
two pixels one upon the other so i think this might cause this problem.
Is there a way to read the point as point instead of reading a comma?

thx, br klemensl

--
SEQIS Spam & Virus Agent: Message ok

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

[PATCH] make ocraf 0.17 build for me: add missing include files, and missing std:: prefix

2008-05-27T05:44:10Z

Hi,

this makes it build on x86_64, Linux 2.6.

Note that I've added an include to a header file that didn't have any includes
in there (for FILE). If you meant that file to not include anything
you'll have to
include <cstdio> in the cc files that need it instead.

While there - are there any plans to make ocrad into a library that can be used
by other programs ?

Thanks

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

ndOCRAD in Blind Assistant

2008-05-11T08:55:26Z

Good morning, I'm Filippo Battaglia from
Italy. I'm the responsable for the development
of Nanodesktop Project and Embvision
Project at Visilab Research Center -
University of Messina - Italy.

We're created a version of OCRAD that is
compatible with Nanodesktop system.

We want to integrate it in our Blind Assistant
program, but we have some troubles.

The performance in recognition are quite
poor. Can you give to me any informations
about the optimization of the engine ?

We are using a OpenCV adaptive thresholding
before passing the cleaned image to OCRAD
engine. But it seems to run bad with Sans Serif
Font.

Can you help me with some tips ?

Thanks
Filippo Battaglia
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Ocrad cannot recognize text in png format file

2008-04-18T04:02:23Z

Hi,

I'm beginner user of ocrad.

Ocrad cannot recognize text in png format file.

The text in png file is as follows:

Filesystem 1K-blocks Used Available Use% Mounted on

The result of invoking command:

pngtopnm filename.png | ocrad

is the following:

|_n__|_n_ O___oO -_ o_

Can you tell me how I need to "force" ocrad to proper recognize text which is "coded" in my png file.

Jerzy
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Ocrad cannot recognize text

2008-04-17T05:30:20Z

Hi,
I'm beginner usr of ocrad.

Ocrad cannot recognize text in pgm format file.

text in png file:

udev 257376 52 257324 1% /dev

The result of invoking command:
pngtopnm filename.png | ocrad
is the following:

udev zs737s sz zs73z4 l?/o/dev.

Can you help me please ?

Jerzy
_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: after installation: ocrad binary not found

2008-04-04T02:47:59Z

Hello! I have the same problem after instalation. Can someone help?
Thanks

Detlef Seinecke wrote:

Hello, after the first installation of your software, there is a mistake
device, that the software can use an ocrad binary. What shell i do?
--
Detlef Seinecke <detlef112@gmx.de>

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: 0.18-pre2

2008-04-01T10:23:09Z

Uwe Dippel wrote:
> Does ocrad allow to set a 'dictionary' of valid characters?

No. The closest thing to your needs that ocrad currently provides is
"--charset=ascii".

Regards,
Antonio.

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad

Re: 0.18-pre2

2008-03-28T20:22:40Z

On Sat, Mar 29, 2008 at 8:08 AM, Antonio Diaz Diaz <ant_diaz@...> wrote:

> > Any good reason, why you tried scaling at a factor of 2?
>
> Well, it is the first thing I try when I find small or not well defined
> letters, because many times improves things.

Interesting, thanks. The scaling by 2 gives a number of better
results. Unfortunately, the scaling done by ocrad has rendered a few
'l' (lower case L) into '}'. So I fell back to pbmpscale (N=2), which
seems to do a good job. But scaling really *is* a great advantage!

Does ocrad allow to set a 'dictionary' of valid characters? What I
mean, is like valid characters are uppercase, lowercase, digits,
punctuation, alphanumeric, or passing a string containing all valid
characters? In some cases, this can be used to increase the accuracy
as well. Like in our application, we use low resolution, but we only
need alphanumeric, '.', '_','-'. Any ambiguity could fall back to one
of these.

> > And still another one: is there a chance to scale with different factors
> > for x and y? I am asking, since we are doing OCR on faxes, and 'normal'
> > at faxes means 204x98 dpi.
>
> Ocrad can't currently scale differently in both axes, and I can't
> remember a case where this could help. But perhaps you could fin one. :)

See above. I have now inserted pnmstretch -yscale=2, followed by
pnmsmooth. It is still not optimal, but at least the ratio is correct.

Uwe

_______________________________________________
Bug-ocrad mailing list
Bug-ocrad@...
http://lists.gnu.org/mailman/listinfo/bug-ocrad