[opendtv] Re: Image quality
- From: Tom Barry <trbarry@xxxxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Sun, 13 Nov 2005 14:03:09 -0500
Mark -
I get confused in going back and forth between computer and TV terms so
I may be saying something stupid here. But in my own playing with
various sharpening algorithms I always eventually run up against the
point where sharpening or whatever starts to just amplify quantization
or other noise. That is, it works well for high bit precise source but
not very well for whatever 8-bit previously compressed source I might be
playing with. I realize deconvolution is much more complex but the same
principles likely apply.
It appears you equate limited dynamic range with limited MTF at higher
spatial frequencies. But that limited MTF can come from many sources
and one of those is probably adjacent pixel averaging caused by your
optical limitations. And it is that last factor we would try to correct
if we got the signal in the camera before too much other noise was
introduced.
Am I grossly misunderstanding something here? Or are you just
suggesting all the info is all too badly mangled from the start to do
much with it? At the very least it seems any downsampling method should
take into account any known optical distortions.
But I am just speculating on a Sunday afternoon without either much
thought or knowledge in this area.
Thanks for your patience.
- Tom
Mark Schubin wrote:
> Tom Barry wrote:
>
>
>><>Still, it seems it would be a mathematically simpler problem to just
>>deal with the distortion caused by limited optics.
>
>
> If you know the distortions AND you have enough dynamic range, sure.
> But the reason a crew was sent to repair the Hubble telescope instead of
> simply sticking a computer on the output is that sometimes you don't
> have enough dynamic range.
>
> In Larry Thorpe's paper presented at SMPTE on Thursday, he showed charts
> of camera-limited MTF for various formats. The 2/3-inch 1920 fell to
> about 45% MTF at 872 TV lines per picture height (30 MHz, 1550 lines
> across). That's a little more than one bit down from 100% MTF. The
> 1/2-inch 1440 looked to be about 25%, or two bits down. The 1/3-inch
> 960 looked to be about 6%, or about four bits down. Another one of his
> slides compared the same lens configured to create 2/3-inch, 1/2-inch,
> or 1/3-inch images. At 872 TVLPPH, the 2/3 was about 75% MTF, the 1/2
> was about 55%, and the 1/3 was about 28%. That's another big chunk of
> dynamic range lost.
>
> TTFN,
> Mark
>
>
>
> ----------------------------------------------------------------------
> You can UNSUBSCRIBE from the OpenDTV list in two ways:
>
> - Using the UNSUBSCRIBE command in your user configuration settings at
> FreeLists.org
>
> - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
> unsubscribe in the subject line.
>
>
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
