Quote:
Originally Posted by DiscoDuck
I was pondering the other day whether it was worth using some data I grabbed when the moon was up to add to data I had from a moonless sky, and was just doing a quick back of the envelope calculation when I saw this thread.
You can use essentially the same calculation to decide which of your subs to throw away Rodney, ie how bright does the background have to be in an image before using that image in your stack would make the stack’s final SNR worse?
With some (perhaps oversimplifying!) assumptions, it seems that the answer is that if the background brightness of the new image is less than twice the background brightness of the majority of the images, then it’s beneficial to use it. (This is if you have a lot of images and are only looking at using a few extra bright ones – the more general case is the one I was originally looking at, where you can be adding quite a few images, and there you can get away with using even brighter images).
Note: by brightness I mean average ADU above the dark frame –not sure what that would be in your case Rodney.
|
I guess the issue Paul is that it comes down to the weighting that one chooses to allocate against SNR and where you might set that threshold between the keep or throw in the bin. Any increase in noise from what ever source is going to decrease SNR assuming the signal level remains constant. I am in the fortunate position with my own observatory that I can be less precious about keeping subs that are less than optimal. Alternatively if one has to travel to a dark site and set up just for the weekend (which I have done previously) then one tries to hang on to as many photons as possible.