Residual artifacts and stars after running Starnet

A

akoenig

Member
Hi everyone,
I hope someone can help me on this issue.

I'm working on a SHO version of SH 2-190 (the centre of the Heart Nebula. The details are as follows:

  • Scope: TSRC8C
  • Camera: ZWO ASI294MM Pro with a reducer and 2x2 Binning
  • Focal Lenght: 1056mm
  • Number of subs: approx. 100 x 240Sec per filter
  • Software: SGP and PixInsight (latest versions)
  • Process used: WBPP
  • Drizzle integrated (factor 2)
I combined the subs to SHO after the usual Non-Linear processesing and then followed by the stretch. I then try to seperate the starts from the rest of the content using StarNet and "Create star mask" ticked and the stribe set to 128, 64 and 32. I don't really want to go down to 16 or 8, as the process will, even on my Xenon Workstation with 64GB memory, take between 24 and probably 48 hours and I'm not sure the result will be any different.

The result with a Stride settiung of 32 can be seen in the attached document. As you can see there are square artifacts where the larger stars were and almost all of the smaller stars can still be seen.

This is the drizzled version, the un-drizzled version looks worse.

Any ideas as to what I'm doing wrong? Any suggestions?

Thank you for your time and effort.

Regards,
Allen
 

Attachments

  • After_Starnet.JPG
    After_Starnet.JPG
    195.8 KB · Views: 156
I can only tell you that, for me, Starnet also leaves stars and star artifacts pretty much every time I use it. Usually when I eventually add the stars back into the image, the artifacts are not so noticeable. I presume the software is just not perfect.
 
Do your star profiles look good before removal? If not Starnet will have problems recognising them as stars.

Also you could try using Starnet on individually integrated and stretched S, H, and O subs. This might show up one of the channels as being the root of the problem.

Gordon
 
i agree with gordon - sometimes it does better on a per-channel basis.

regardles starnet should do a much better job than what you are seeing, so there must be some peculiarity of your image. this image is from an AT6RC and starnet did pretty well -



rob
 
You must log in or register to reply here.
Back
Top