Hot Pixel Suppression Question

Somewhere to ask questions about the best way to use SharpCap
Forum rules


If you have a problem or question, please check the FAQ to see if it already has an answer : https://www.sharpcap.co.uk/sharpcap-faqs
Post Reply
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Hot Pixel Suppression Question

#1

Post by MarMax »

I've been trying to optimize my workflow for flats and was wondering how Hot Pixel Suppression (HPS) works in relation to an applied Master Flat. This is my general understanding of flats:

Assumptions:
(1) flats are always applied with a complimentary dark
(Q) Does the new SharpCap "hot pixel suppression" feature in-lieu of a dark comply with this assumption?

(2) flats/darks/lights taken with the same image train (filter, etc.) and at the same temperature

(3) flat light source is noise free, linear, smooth, reproducible, and uniform

Flats "best practices":
- low gain
- flat offset = bias/flat-dark offset
- exposure 1-4s

I've used the HPS feature for two EAA sessions and it seems to work nicely. I've only found a few hot pixels doing some post-peeping. But I'm still wondering about the importance of the flat/dark combination and if it's best to apply a Master Dark vs. using HPS?
User avatar
admin
Site Admin
Posts: 13176
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Hot Pixel Suppression Question

#2

Post by admin »

Hi,

currently the hot pixel removal in SharpCap is an alternative to subtracting a dark frame, so if you activate HPR then you will get less hot pixels, but not the background level removal that a dark frame will usually bring.

Now, providing you (a) are using flats that you captured with either bias or dark subtraction endabled and (b) don't have significant thermal noise that you need darks to remove, then flats should still work pretty well. That's because when you capture a flat with a bias or dark in SharpCap, the background level in the bias/dark frames is recorded into the flat file name as the 'offset=x.xxx%' bit at the end. When you reload the flat and start using it, SharpCap will use that offset level if there is no dark frame in use to subtract out the image floor pixel values from the light frame before applying flat correction.

So, overall, I think the answer to your question is that the new hot pixel removal is not the thing that is helping you here, it's the fact that the SharpCap flat correction has a backup for removing the offset level if dark frames are not used.

cheers,

Robin
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Re: Hot Pixel Suppression Question

#3

Post by MarMax »

Thank you for the reply Robin.

As a follow up, for someone doing only EAA, do you feel that HPS with a Flat (with bias subtraction) will provide results similar to a Flat (with bias subtraction) and a Dark?

Or maybe a better way to ask this is if you were doing an EAA session and not in a rush, would you take Flats and Darks or just Flats and use HPS?
User avatar
admin
Site Admin
Posts: 13176
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Hot Pixel Suppression Question

#4

Post by admin »

Hi,

I think it really depends on the camera - I did some experiments with data from a modern IMX533 based camera and also from a somewhat older 294 based camera. Darks have less value with the newer 533 camera than with the older sensor. My conclusions are based on a couple of approaches

1) Process data with/without darks and see what noise levels you get in the final image

2) Split the dark frames into two halves to make two 'master' dark frames. See how much of the variation in one is explained by the variation in the other.

Unfortunately the bit of paper I had jotted down the numbers on has gone missing, so I am doing this from memory, but certainly in the second case there was a big difference between the two camera models - the 'A' dark explained a high percentage of variation (70-80?) in the 'B' dark for the 294, but significantly lower (20-40?) for the 533.

In the end though, maths only takes you so far - the real proof is in the image. If you image without darks and get problems like walking noise being very noticeable then you need to start taking darks again.

cheers,

Robin
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Re: Hot Pixel Suppression Question

#5

Post by MarMax »

OK, thank you for the follow up Robin. I like your final point about walking noise. That helps me know what to look for. It seems like EAA is the poster child for walking noise. I rarely use 30 second subs and typically I'm at 15 seconds or less. If I'm using HPS and a flat and don't see any walking noise then I know it's probably as good as it would be with a dark.
User avatar
admin
Site Admin
Posts: 13176
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Hot Pixel Suppression Question

#6

Post by admin »

Hi,

another thing I forgot to mention is that it also depends on how hard you stretch your final images. If you have a faint target you may well stretch more than usual, which can bring defects like walking noise that were previously not a problem into view.

cheers,

Robin
Post Reply