Author Topic: Multiple Masks  (Read 2060 times)

Offline CRWPhoto

  • Newcomer
  • Posts: 16
Multiple Masks
« on: 2016 July 05 05:35:41 »
I'm not sure if there is a way to already do this...

I found myself editing the veil nebula and really wanted to be able to apply two masks to an image to isolate the nebula.  A really simple way to do this would be to create a Lum mask and clip in HT.  Apply to image, then create a star mask and apply that to the image to further mask off the areas you dont wan to touch. 

I'm guessing that this could be achieved with pixel math, but that is beyond me at this time.  It would be so simple (UI) to drag and drop the first mask, have that tab visible, then drag and drop the second mask.  You could then independently remove and invert those masks depending on what you want to do for edits.  I find myself wanting to do this more and more frequently.  It sounds simple to do, but maybe it is not that simple when executing this in the program. 

I'm not a programmer, and only an amateur user of PI so simple execution of a concept like this would really be helpful.  Does this already exist?  Kudos to the programmers on this program... it is just awesome!  Thank you!

Offline pfile

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 4729
Re: Multiple Masks
« Reply #1 on: 2016 July 05 09:33:50 »
i don't think it exists and as you mention currently the only way to do this is to use pixelmath.

it's not too bad though, you can multiply one mask by the other ("*" operator) and if need be invert one of the inputs to pixinsight (~ operator), so maybe ~mask1*mask2.

or you can say $T*mask2 and drag the pixelmath triangle onto mask1 which will apply *mask2 to the target ($T is a special variable that means whatever the target image is).

or perhaps you can do max(mask1,mask2) to pick up the brightest pixels from either mask.

rob

Offline CRWPhoto

  • Newcomer
  • Posts: 16
Re: Multiple Masks
« Reply #2 on: 2016 July 09 13:29:04 »
Thanks pfile.  I think that will work perfectly for me.  That is pretty simple and not nearly as daunting as I was thinking it would be.  I appreciate your reply!