Stacking algorithms

Discussion of using SharpCap for Deep Sky Imaging
Post Reply
jpberger
Posts: 4
Joined: Fri Oct 06, 2023 3:14 am

Stacking algorithms

#1

Post by jpberger »

Hi there, with my QHY600M, when I use the Sigma clipping algorithm, I have plenty artefacts in my image like rain drops. Here a image of 7 stacks of 300 sec. Click on the image and after use the function ''max resolution''.

https://www.astrobin.com/q2aj49/

With the Max Value algorithm, I have more noise of course, but no artefacts. Here a image of 7 stacks of 300 sec.

https://www.astrobin.com/qjpgjj/

Personally, I prefer Sigma clipping since airplane traces disappear with the average value or are ignored after 5 subs. Also, less noise to treat in the post-processing.

Is it a bug with the Sigma clipping algorithm?

Best regards.
User avatar
admin
Site Admin
Posts: 12970
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Stacking algorithms

#2

Post by admin »

Hi,

the Sigma clipping is limited in the amount it can do to correct these sort of artifacts by the fact that it has to run on data as it arrives rather than being able to process all images at once as you would in conventional processing... In particular, the sigma clipping cannot be applied at all for the first few frames (default 10?) while some statistical information about the brightness of the image is being accumulated. I'm not sure how many individual images went into your stack, but it's possible the short trails are coming from those initial frames...

To me, the short trails look like hot/warm pixels being spread out into lines by the gradual drift of the image across the field of view. SharpCap is correcting for the image drift in stacking, but that leaves the warm pixel trails in the stack. The best thing to do is to get rid of those pixels before the image is stacked. This could involve

* Using dark frames if you are not already or
* Using SharpCap's hot/cold pixel removal option in place of a dark frame

Cooling the camera further may also help reduce the effect.

The next approach beyond that is to use dithering during capture - small random movements of the mount between frames. This helps break up the highly noticeable pattern you are seeing and spread the excess brightness of those warm pixels on the sensor across the stacked image in a much less noticeable way.

cheers,

Robin
jpberger
Posts: 4
Joined: Fri Oct 06, 2023 3:14 am

Re: Stacking algorithms

#3

Post by jpberger »

Thanks for your quick reply Robin. I did use dark. A new set of darks since you explain me that I have to be in a still mode before the stacking. But it changes the exposure size. I will try at -20 degrees.
User avatar
admin
Site Admin
Posts: 12970
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Stacking algorithms

#4

Post by admin »

Hi,

Ok, good luck with the new try. Do make sure all settings are the same between dark and light frames for best effects, and take plenty of dark frames when making the master dark to avoid noise in the master dark being part of the problem.

If you are already at -5 to -10 then going cooler will have limited effect with modern CMOS cameras (unlike older CCDs which benefit more from extreme cooling).

cheers,

Robin
Post Reply