[comp.ai] Optical Character Recognition

ap1a+@andrew.cmu.edu (Andrew Balen Philips) (09/28/88)

I am a student at Carnegie Mellon, and I am working on a project
to perform Optical Character Recognition.  I have the hardware to
get text information into files.  What I need is the software to process
those files.  Are there any programs out there that I could use, or
does anyone have information on people I could contact?

Your help is appreciated.  Thank you.
Please send mail directly to me.  If there is enough requests, I will
compile and post the information to this bulletin board.

Andy Philips
ap1a+@andrew.cmu.edu

jbn@glacier.STANFORD.EDU (John B. Nagle) (09/28/88)

     Kurtzweil and Palantir both build special-purpose machines to do
general multi-font character recognition.  Their algorithms are proprietary.
The Palantir unit is said to have about 300 MIPS of computational power
inside.

     Doing this badly is easy.  Doing it well is very, very hard.

					John Nagle

root@vortex.COM (Superuser) (09/03/90)

Query: Is there any OCR software for Unix (running under X is not
necessarily required) which will take a bitmap, GIF, etc. input
file and attempt to convert the text to ASCII?  Such programs
exist for DOS and the MAC... how about Unix?  Thanks.