Author Topic: Background Neutralization fails in my first three image attempts  (Read 1596 times)

Offline tstephens

  • Newcomer
  • Posts: 25
    • View Profile
After an RGB combination and application of linked auto STF, I get a strong background color (different in my three image sets). I open the Background Neutralization process and select a background preview that should be neutral and apply it to these images. Nothing happens. The linked auto STF still has the same color bias. When I look at the Histogram Transformation you can clearly see the bias. When I do the NGC 1808 example, it works for me. It just does not work with my images. What am I doing wrong? I would attach a set of .fit files but does not appear to be an option with this post engine.

Offline pfile

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 4691
    • View Profile
Re: Background Neutralization fails in my first three image attempts
« Reply #1 on: 2013 October 25 12:34:56 »
you'd have to upload the fit to dropbox or something.

anyway you need to tweak the upper limit to be something reasonable for your image. run your mouse over the background and look at the readout; pick an upper limit that's just slightly larger than the largest of the 3 channel values.

rob

Offline tstephens

  • Newcomer
  • Posts: 25
    • View Profile
Re: Background Neutralization fails in my first three image attempts
« Reply #2 on: 2013 October 25 12:38:29 »
I will have to learn how to use the dropbox, however, your suggestion has already paid off. Increasing the upper limit above the largest number seen when going over the background with the mouse has allowed me to achieve a successful background neutralization.

Offline pfile

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 4691
    • View Profile
Re: Background Neutralization fails in my first three image attempts
« Reply #3 on: 2013 October 25 14:09:24 »
great! you can probably be more scientific about it by defining your preview and then using the statistics process to get the mean value for each channel. i pretty much never do that though.

also it's worth using the "region of interest" rather than pointing the process at the preview; if you decide to reprocess later (say, more subs or something) then you can re-use a saved process icon on a new image without first defining a new preview on the new image. needless to say it might be tedious to define the exact same preview twice, but after using ROI you don't need to worry about it anymore.

rob