Re: Interesting way to image with a cheap mount


Benoit Schillings
 

well, this is basically an information theory problem. blurred images
do contain less information from a Shannon IT standpoint. the basic
info is more or less the resolution convolved with the signal to
noise, and this puts some limits to how well you can invert the
information. Basically, if you do want to deconvolve or anything
similar, the information (entropy) has to stay constant, so you are
trading resolution for signal to noise. So, yes, you could use a lot
of bad data to create some better one, but you cannot create data out
of nothing, and the resolution goes about as the third power of the
initial signal to noise (if I remember well). is this the tradeoff you
want to do ?

-- benoit

On Fri, Sep 10, 2021 at 7:41 PM Dean Jacobsen <deanjacobsen@...> wrote:

I love my FSQ-106 too.
--
Dean Jacobsen
Astrobin Image Gallery - https://www.astrobin.com/users/deanjacobsen/

Join main@ap-gto.groups.io to automatically receive all group messages.