From: Alan Friedman <alan@greatarrow.com>

Date: April 9, 2008 3:09:58 PM MDT

To: Astro_IIDC@yahoogroups.com

Subject: Re: [Astro_IIDC] Mathematics of Stacking

Hi Jim,

I believe that the S/N ratio improves by a factor of 2 to the square of the number of frames used, ie: 100 frames stacked would improve the S/N over 10 frames by a factor of two - to gain another 2x increase in signal quality, 1000 frames would be required. 

best,

Alan

On Apr 9, 2008, at 4:17 PM, jimchung2338 wrote:

Hi all,

I'm giving a talk on planetary imaging for members of the Toronto RASC
group and have always wondered about the math of stacking. This is
what I feel and welcome comments:

1. Stack the sharpest number of frames at a high cutoff like 90%
even if its only a hundred frames since the firewire cameras we use
here are much less noisy than webcams we don't need to reduce noise by
stacking several hundred frames and have sharp details diluted by
poorer frames.

2. If I'm doing a LRGB image I would do #1 for L channel and then
stack several hundred frames at a lower cutoff to reduce the noise
that usually accompanies turning up the gain to image through RGB filters.

3. Is there a mathematically determined number of frames were
additional stacking produces diminishing returns in quality.

Thanks!

Jim