Strange banding artefacts when using WBPP 2.5.1

StuartT

StuartT

Well-known member
I just did an integration and am seeing some ugly banding at the top and bottom of the image. I didn't get this with the previous WBPP. Any ideas?
 

Attachments

  • Capture.JPG
    Capture.JPG
    622.6 KB · Views: 86
John_Gill said:
I think this is from guiding and I don't think you have you cropped the image?
Click to expand...
no. not guiding. This is integrating a set of RGB light frames which don't have this banding present (so it's not about mismatch between channels).
 
clutching at straws but if you combine three subs alone do they align ok? Did you blink all the subs to make sure there are no rogues?
 
scotty38 said:
clutching at straws but if you combine three subs alone do they align ok? Did you blink all the subs to make sure there are no rogues?
Click to expand...
yes, they blink fine. All seem very closely aligned.

This is what one of the individual subs looks like
 

Attachments

  • Capture.JPG
    Capture.JPG
    670.3 KB · Views: 52
Ah ok, when I said to combine three subs I thought you had mono R, G and B..... I guess if they're all colour subs from the same session we can rule out clutching at alignment straws. No flip involved at all????
 
scotty38 said:
Ah ok, when I said to combine three subs I thought you had mono R, G and B..... I guess if they're all colour subs from the same session we can rule out clutching at alignment straws. No flip involved at all????
Click to expand...
sorry. Should have been clearer. These are images from a remote telescope which shoots mono subs with Ha, Sii and Oiii filters, but then delivers the combined images as colour. I then integrated them.

Yes, there was a flip half way through
 
Thanks and now I may be confused.... If they shoot mono and send you a colour image then isn't it already integrated? What am I misunderstanding here?
 
scotty38 said:
Thanks and now I may be confused.... If they shoot mono and send you a colour image then isn't it already integrated? What am I misunderstanding here?
Click to expand...
I agree, it's an odd approach. It's the COAST telescope on Mt Teide, Tenerife. What I mean is that they send each S, H, O sub I order (e.g. a 300sec exposure on each filter) but already combined SHO into a single colour sub. I then integrate all the colour images (think OSC processing - you are stacking individual RGB subs).

They do offer an option to download the individual grayscale S, H, O subs too. In fact, I just did that and integrated them channel by channel, but I still got this odd artefact.
 
Ok, I've just done another WBPP run and I am getting the same light bands at the top and bottom of the integrated result.

Is it possible for me to have my PI set back to the previous version of WBPP?
 

Attachments

  • Capture.JPG
    Capture.JPG
    146.9 KB · Views: 48
Last edited:
Still confused on your explanation of the colour stuff as each S H O isn't colour and not seeing how they make three colour channels for you to stack.

Anyway that aside can you share a few of each of the greyscale SHO subs and I'll have a go too. Can't help thinking it's something odd about what they sent you combined with how they're being processed if you see what I mean.
 
scotty38 said:
Still confused on your explanation of the colour stuff as each S H O isn't colour and not seeing how they make three colour channels for you to stack.

Anyway that aside can you share a few of each of the greyscale SHO subs and I'll have a go too. Can't help thinking it's something odd about what they sent you combined with how they're being processed if you see what I mean.
Click to expand...
sorry if I'm not explaining myself very well. What I mean is that the FITS file they give me is already channel combined (S-->Red, H-->Green, O--> Blue). So it's a colour image. Then I integrate a bunch of those.

Anyway, thanks for your offer of help. Here are four of the colour images.
And here are the corresponding grayscale SHO images they are made from

I never used to have any problems doing this with the previous versions of WBPP. They have changed a lot of things this time (for example the platesolving - not sure what that's doing in there now!)
 
scotty38 said:
They need cleaning up first I'd say
Click to expand...
Sure. I am aware of the white bands of noise along the bottom and right hand edge. This seems to be a problem with the sensor and is on all the images from this telescope. But in the previous WBPP the integrations all proceeded normally (i.e. unaffected by those bands) and I would then simply dynamic crop off the noise bands in the resulting image.

So, I had a theory ... maybe it is the new Linear Defects Correction stage that is being upset by these bands (which would account for why this didn't occur in the previous version of WBPP).

I tested it by disabling LDC and sure enough, the banding doesn't appear in the integrated result.

Mystery solved!

Most satisfying to have gotten to the bottom of that. Thanks for your help :)
 
You must log in or register to reply here.
Back
Top