Re: Handwriting Recognition: See Linux Journal
on
Apple's New Trackpad?
·
· Score: 2
Just a quick note, I've found a better description of this system at http://www.business-linux.at/case/case.uspostal.ht ml
Re: Handwriting Recognition: See Linux Journal
on
Apple's New Trackpad?
·
· Score: 1
"The United States Postal Service deployed over 900 Linux based systems throughout the United States in 1997 to automatically recognize the destination addresses on mail pieces. Each system consists of 5 dual Pentium Pro 200MHz (PP200) computers and one single PP200 all running Linux."
For info on the US postal service handwriting recognition why don't you look at an old copy of Linux Journal?
I forget the excat issue number but it was within the last year or so, it could also possibly been an "on-line only" article. Linux Journal subscribers will know what I mean.
They had an article describing how it was developed, its a cluster of computers over a small dedicated lan.
One computer is in charge of doing high resolution scanning of the envelopes, and others are in charge of actually doing the recognition, or something like that. Apparently it randomly saved some of the scans to a harddrive for latter use. This is how the developers "tuned" the system to make it recognize things better.
Slashdot Mirror
Just a quick note, I've found a better description of this system at http://www.business-linux.at/case/case.uspostal.ht ml
For info on the US postal service handwriting recognition why don't you look at an old copy of Linux Journal?
I forget the excat issue number but it was within the last year or so, it could also possibly been an "on-line only" article. Linux Journal subscribers will know what I mean.
They had an article describing how it was developed, its a cluster of computers over a small dedicated lan.
One computer is in charge of doing high resolution scanning of the envelopes, and others are in charge of actually doing the recognition, or something like that. Apparently it randomly saved some of the scans to a harddrive for latter use. This is how the developers "tuned" the system to make it recognize things better.