Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Getting started with mosaics

  • Please log in to reply
4 replies to this topic

#1 mdavister

mdavister

    Messenger

  • *****
  • topic starter
  • Posts: 453
  • Joined: 10 Jul 2007
  • Loc: Pennsylvania

Posted 27 September 2022 - 08:13 AM

I've been wanting to get started in trying mosaic imaging, but am unsure of a couple of things. First at the point of data acquisition, what does everyone else do? Bounce back and forth between locations, panel 1, panel 2 ... etc., or just do one panel for a bunch of hours, wait for the next available day rinse and repeat? Second, when you get to processing the data what is the general workflow? I have been working with PixInsight, so I would assume that it is calibrate, de-bayer, register, integrate, and then gradient combine?

 

Biggest question surrounds how to not get obvious seams in the image when processing. As one is imaging, the sky is being tracked and entering various degrees of light pollution. With a single panel this background can be mostly compensated for. When a mosaic is put together, if the panels aren't perfectly matched a seam will be visible, or at the least one panel will have a potentially different "look" to it. 

 

I'm sure there are a bunch more questions, just trying to see how deep the water is before diving in.



#2 whwang

whwang

    Skylab

  • *****
  • Posts: 4,240
  • Joined: 20 Mar 2013

Posted 27 September 2022 - 11:14 AM

Mosaic is tricky.  You need nearly perfect flat-fielding and gradient removal to achieve seamless mosaicking.  The latter is highly challenging, or practically impossible, if your images are full of nebulas everywhere without many clearly identifiable "background" area for the gradient removal.  There are tools that help to workaround, like PI's photometric mosaic script, or APP's global approach of gradient removal.  PI's gradient merge mosaic function also attempts to tackle the gradient problem, but unfortunately it has mixed reputation.  To me, anyone who can routinely achieve seamless mosaics with PI must be a black-belt level master.  For beginners, may would recommend APP for mosaicking.

 

For simple 2-panel mosaics, I always try to get equal amount of integration on each of them in the first night.  This way, I have a full mosaic to work with after that night.  An obvious advantage is that I will have something to show people. Another advantage is that you will be able to access the quality over the whole mosaic and decide later if you want to deepen it.  If you are a beginner, I would suggest this approach.  Later when you know very well what you are doing, you may go deep right away into a panel, and then complete the panels one by one, over many nights.



#3 Eric Horton

Eric Horton

    Ranger 4

  • *****
  • Posts: 332
  • Joined: 04 Dec 2011
  • Loc: Deltona, Florida

Posted 27 September 2022 - 11:16 AM

It's going to depend on your capture software. Programs like voyager, sgp, and I believe Nina all have mosaic abilities where you can frame and acquire your panels. Since you are on pixinsight there are many utilities for dealing with the data and not have the seem. There are tutorials on lightvortexastronomy.com.

#4 Rasfahan

Rasfahan

    Vanguard

  • -----
  • Posts: 2,107
  • Joined: 12 May 2020
  • Loc: Hessen, Germany

Posted 27 September 2022 - 02:13 PM

With LocalNormalization in WBPP you‘re off to a better start than only a few months ago, greatly reducing the gradient problem. Photometric Mosaic Script seems to be nice, too. But do try APP for mosaics: Integrate each panel separately, then combine them - it works quite well without needing a lot of input. During data aquisition I try to gather one panel per hour, then switch to the next. That way I have data in all panels even when the weather changes on the next night. If you have reliable clear skies I‘ld try to catch the panels at the same altitude.


Edited by Rasfahan, 27 September 2022 - 02:43 PM.

  • mdavister likes this

#5 mdavister

mdavister

    Messenger

  • *****
  • topic starter
  • Posts: 453
  • Joined: 10 Jul 2007
  • Loc: Pennsylvania

Posted 27 September 2022 - 08:16 PM

With LocalNormalization in WBPP you‘re off to a better start than only a few months ago, greatly reducing the gradient problem. Photometric Mosaic Script seems to be nice, too. But do try APP for mosaics: Integrate each panel separately, then combine them - it works quite well without needing a lot of input. During data aquisition I try to gather one panel per hour, then switch to the next. That way I have data in all panels even when the weather changes on the next night. If you have reliable clear skies I‘ld try to catch the panels at the same altitude.

That is some useful info. I will have to look into the sequencers I use and see what tools they have. SGP has been my main one, I'm starting to dabble in NINA to just see what's different. 

 

I don't get as many clear skies as I'd like, usually its a good night then a week or two of clouds. By then the moon is at a different phase and casting a different sky glow. I'm guessing it looks like careful processing management and maybe bouncing around between panels is the best option. 




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics