From: hfinney@shell.portal.com (Hal Finney)
To: cypherpunks@toad.com
Message Hash: 8fffeaff79c53a8a66e9745eae868a9f37468a4d2cff5039d66c2c8858db963f
Message ID: <9310221617.AA00624@jobe.shell.portal.com>
Reply To: N/A
UTC Datetime: 1993-10-22 16:18:17 UTC
Raw Date: Fri, 22 Oct 93 09:18:17 PDT
From: hfinney@shell.portal.com (Hal Finney)
Date: Fri, 22 Oct 93 09:18:17 PDT
To: cypherpunks@toad.com
Subject: Re: Canon copiers
Message-ID: <9310221617.AA00624@jobe.shell.portal.com>
MIME-Version: 1.0
Content-Type: text/plain
The thing that makes me skeptical about this copier story is this: the
money could have any orientation and position on the page. It would take
a large amount of computing power to look at an image and deteri\\ determine
whether there is a certain bitmap anywhere on it. Look at how poorly
optical-character-reader technology does now, and that is when it knows
how the letters are oriented. Adding random orientations would make the
problem far worse. And, will it look at the whole dollar bill, or just
pieces of it? It has to respond to all the different denominations, too.
Plus, if it makes a mistake and permanently locks up the customer's machine,
the manufacturer is going to have a big problem.
In short, I'm pretty sure that there is a lot of disinformation going around
designed to scare people away from trying it.
However, there is another possibility, which is to look at hte color of
the bills. This does not take so much processing power, although bills
may vary somewhat in color. Years ago, I'd read that these machines would
not accurately reproduce the color of money. Perhaps today they will turn
that color to black.
Hal
Return to October 1993
Return to “Matthew J Ghio <mg5n+@andrew.cmu.edu>”