Re: How does projection magnification in astrophotos change the Airy disk size?

You can however use phase diversity (extra focal images) to estimate
the PSF with out seeing provided that the exposure is much longer
than the seeing time scales. I will admit that this is not in the image plane as you two seem to be chewing on. It is however using starlight.

On an aside I knew an old double star observer, well published, that claimed that he knew
the psf of each telescope he used (visual) and could then, see very close pairs.

a sort of out of focus
Dan


Chris L Peterson wrote: 
On 25 Apr 2005 20:44:22 -0700, "nick" <vladis.2@xxxxxxxx> wrote:


Not clear what is "normal" (or abnormal?) distribution. Aside the more
or less typical formalistic description, the PSF in imaging usually can
be
approximated by Gaussian functions due to disappearance of ring
structure
(Gaussian functions do not form successive minimas and maximas,
and purely approximate the bright rings structure).



The reason the PSF approximates a Gaussian is because if you convolve
the Bessel function that describes the diffraction (and includes ring
structure) with the statistical position shifts caused by tracking
errors and seeing, the resulting function assumes a shape that is very
nearly Gaussian. In the absence of these shifts, you would not use a
Gaussian to describe the PSF.



Limiting "resolution" to that of a pair of point objects images, for
the sake of
simplicity (very much needed here, indeed), resolution can be described
as producing an image with such a pair fully or partially resolved.
Therefore,
it is not directly determined by the PSF, rather by the property of the
detector



It has nothing to do with the detector, unless you are undersampled. We
are talking here about _measuring_ the actual PSF (by looking at a
stellar profile) in order to _calculate_ the maximum possible resolution
of the image. This has nothing at all to do with any actual features in
the image. That is, there may actually be nothing in the image (aside
from stars) that actually contains high frequency components. The
measurement just tells you the best resolution you can have. You can't
judge this visually. If you have a pair of close stars, just changing
the black and white points can determine whether they _appear_ as
separate or joined. But if you have measured the PSF, you know right
away the degree to which the pair is resolved.




The FWHM is determined by the PSF, which is in turn
directly dependant on the wavelength in its limiting (diffraction)
form.



When imaging, the PSF is almost unrelated to any diffraction effects.
The PSF is primarily the result of seeing, resulting in a FWHM several
times larger than the Airy disk.



I'd love to see an example of an image where the closest resolvable
stars and the FWHM of the stars in the image are unrelated.

Who seeks, will find. 


I'll take that to mean you can't provide one.

I'm glad you've found a system of assessing resolution that you are
happy with. Personally, I'll stick with the system used by every other
imager in the world.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com


.

Relevant Pages
Loading