On Tue, 11 Jan 2000, Eric Anderholm wrote:
>
> Greetings:
>
> I have a functional machine print ocr engine I have built and am
> wondering if it would be worth while to open source it and release it to
>
> the net. It is not a "single font" trained engine but can do multiple
> fonts and proportional and is completely trained.
>
> Before I go to any trouble of packaging it up with a good auto
> configure script I was wondering if any one is interested in it and more
>
> importantly if anyone wants to work on the back end of it. That is the
> n-gram dictionary lookup.
>
> If your interested in it, and/or may be interested on working on the
> back end please email eja@altavista.net.
>
I am sure I am not alone in thinking there would be great interest
in this project, I for one have been waiting a fair time for a decent OCR
package to be available for Unicies at a reasonable cost. Yes there are a
few commercial libraries/drivers available (for a price) but non have the
flexibilty and adaptabilty that an OS OCR package would bring. Think of
all the other potentially interesting projects that could sping up from
it. Anything from fax to email, or more interestingly proper hand
recognition for the up and coming embeded Mobile Linux/ linux handhelds.
So yes.. If you are willing to release your work and would like
others to build and improve on it I am all in favour of it. I for one
would certainly take an interest in it.
Rooster - A friend.
"Anyone's friend in particular or just generally well disposed to people?"
HHGTTG.
CUT - Campaign for Unmetered Telecomunications:
Bring the best to all affordably. http://www.unmetered.org.uk
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com
This archive was generated by hypermail 2b29 : Sun Jan 30 2000 - 11:49:19 PST