A couple of people have asked why one might want to do this.
The common name for what I am trying to do is adaptive technology.  As a
person who likes to experiment with hardware and who is also blind, I would
like to figure out a reasonably cheap way to read LED and maybe even LCD
displays without having to cut up the device being read.  In many cases,
that is not even practical these days due to the level of integration.

        The idea would be to put the pickup array very close to the display
and then look for the presence or absence of light at key areas to determine
which characters were being displayed.

        One person suggested a CCD camera which may turn out to be the
best idea.  Decoding a full-fledged video signal may be the most complex part
of the project, but it will most likely be flexible enough to adapt to
differing display styles.

        This whole idea is just a continuation of an age-old foot race between
newer technologies and making them accessible to more users.  When I was
first learning about amateur radio in the early sixties, one of the slickest
adaptations was a very cheap and simple voltage controlled oscillator circuit
that got its control voltage from the voltage across a meter movement such
as found on a VU meter or the plate current or grid drive meters on a
transmitter.  The meter was "read" by moving a slide switch back and forth
while turning a potentiometer until the tone heard from the external voltage
matched the one heard from an internal reference that was varied by the pot.
There was usually a Braille scale that you could read and know where the
meter pointed by that method.

        Anyhow, if I can find some easy way to read digital displays, lots
of neat things will be possible.

Martin McCormick WB5AGZ  Stillwater, OK 36.7N97.4W
OSU Center for Computing and Information Services Data Communications Group