RE: Digicam

["gary" <gsellani@xxxxxxxxxxxxx>]

I have to disagree with you on the first count (sampling). Think about this
a bit more before you reply. Input aliasing should not be part of the
discussion. Period. I am focusing (heh heh) on the sample "width", which
leads to the pixel width. Now it is true that imaging does not have a
pre-filter, but so what. Dogs have tails, but this is not relevant either.

Do I need to loan you a copy of Oppenheim and Schafer?

Your comments on the analog of post filtering are not lucid enough to give
comment.

> -----Original Message-----
> From: owner-scan@leben.com [mailto:owner-scan@leben.com]On Behalf Of
> Kennedy McEwen
> Sent: Friday, June 14, 2002 1:52 AM
> To: scan@leben.com
> Subject: Re: Digicam
>
>
> In article <FAEBJHPJNNGCAGDNGLNPIEIHEIAA.gsellani@accesscom.com>, gary
> <gsellani@accesscom.com> writes
> >Those familiar with audio sampled data systems know that the
> sample time is
> >infinitely small. Equating this to imaging, the pixel should not
> have size,
> >but rather a sample density.
> >
> No.  Whilst the audio analogy certainly has a very short sample
> acquisition period, the signal is band limited by the pre-filter - which
> has the same effect as convolving the time domain waveform with an
> aperture of a specific shape and size.  (multiplication in frequency
> space corresponds to convolution in time).  The finite aperture of the
> pixel or spot performs a similar presampling filter function in image
> sampling, and the shape and size determine the exact spatial bandwidth
> that is samples.
>
> >Going back to the audio analogy, if you take the  sample
> impulses and play
> >them back on a DAC making staircase waveforms, the high
> frequency response
> >will be incorrect. The nature of the staircase implies a sinc filter was
> >used, so the playback must incorporate an inverse sinc filter. Since the
> >image sampling is not infinitely small, I wonder if the use of an unsharp
> >mask is analogous to the inverse sinc filter.
> >
> No - an unsharp mask has a completely different spatial frequency
> response from a sinc or inverse sinc, so it isn't doing anything similar
> to it.
>
> Remember that in order to see (or hear) the staircase in audio you must
> examine the waveform in sufficient detail to see individual samples as
> they are converted by the DAC.  In imaging, this is the same as seeing
> teh individual pixelation.  So the equivalent of filtering the
> transition between filters is to resize the image with pixel replication
> and then apply an inverse sinc filter of the appropriate size.  As it
> turns out, for zoom ratios up to around 2-3x, this is closely
> approximated by bicubic interpolation, not unsharp masking.
>
> Also remember that when viewing at normal size, the CRT spot or LCD
> pixels perform a similar spatial filter on the final image.
> --
> Kennedy
> Yes, Socrates himself is particularly missed;
> A lovely little thinker, but a bugger when he's pissed.
> Python Philosophers
> -
> Turn off HTML mail features. Keep quoted material short. Use accurate
> subject lines. http://www.leben.com/lists for list instructions.
>

-
Turn off HTML mail features. Keep quoted material short. Use accurate
subject lines. http://www.leben.com/lists for list instructions.