On Fri, 30 Sep 2005, [EMAIL PROTECTED] wrote:
You are wrong. The problem you get without oversampling is the
information you dont have, and that is the information between the
pixels. The lacking information gives multiple solutions and those
produce several harmonic spatial frequencies. That's why DA converters
etc use a lot of tricks to avoid folding, to reduce the number of
unwanted solutions. Oversampling is a very simple approach, and is what
you get if you have more pixels than necessary.
I know it sounds strange, but I haven't got the time so you'll have to read
some Nyquist on you own.
DagT
if the lens already bandlimits the signal, it's already doing the
antialiasing. more pixels won't render any more information nor prevent
aliasing any better.
Herb....
Another something with images is the "Kell factor." I believe
it's an attempt to quantify the perceived sharpness of a sampled image.
Representing the frequencies is important, but the phase of those
frequencies will affect perceived sharpness. I've run into that while
working with video captures.
-Cory
--
*************************************************************************
* Cory Papenfuss *
* Electrical Engineering candidate Ph.D. graduate student *
* Virginia Polytechnic Institute and State University *
*************************************************************************
