Author Topic: Masked stretch and big star appearance  (Read 374 times)

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Masked stretch and big star appearance
« on: 2019 September 18 02:39:11 »
Hi everyone,

I am working on a narrowband project and the best stretching method for obtaining nebular detail and equilibrium (O3 is really faint) in this cas was Masked Stretch. However the bigger stars in the image have completely saturated circular white core with a net separation from the rest of the star (similar to the deconvolution artefact when the stars are not masked). Is this a known artefact of masked stretch on big stars and are there any setting for obtaining a more gradual tapering of the signal from the center of the star?

Thanx.

Bogdan

Offline dld

  • PixInsight Addict
  • ***
  • Posts: 114
    • View Profile
Re: Masked stretch and big star appearance
« Reply #1 on: 2019 September 18 04:39:36 »
Hello,

The tooltip over the Iterations setting of MaskedStrech suggests increasing the default number (100) of iterations in such cases. Does this work for you?

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: Masked stretch and big star appearance
« Reply #2 on: 2019 September 18 05:26:46 »
I am already using 100 or higher, does not seem to depend on this parameter. I'll try to post an image when I get home, the center of the big stars is saturated (uniform white) and round.

Offline drmikevt

  • PixInsight Addict
  • ***
  • Posts: 112
    • View Profile
Re: Masked stretch and big star appearance
« Reply #3 on: 2019 September 18 07:33:20 »
Depending on how much you are trying to stretch and the saturation of the stars during capture this artifact may be unavoidable.  When this happens to me, I make a mask of the larger stars and apply a very mild Convolution to the cores to smooth them back out again.  It works quite well. 

Mike

Offline Enzo De Bernardini

  • PTeam Member
  • PixInsight Addict
  • ***
  • Posts: 270
  • Resistance is futile.
    • View Profile
    • Astronomí­a Sur
Re: Masked stretch and big star appearance
« Reply #4 on: 2019 September 18 09:41:50 »
You can try this:

MaskedStretch and Stars Cores

Greetings,

Enzo.

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: Masked stretch and big star appearance
« Reply #5 on: 2019 September 18 10:14:32 »
Depending on how much you are trying to stretch and the saturation of the stars during capture this artifact may be unavoidable.  When this happens to me, I make a mask of the larger stars and apply a very mild Convolution to the cores to smooth them back out again.  It works quite well. 

Mike

Thanx that was my plan too. I tried to masked them when I do the Masked stretch but this creates some weird artifacts around the edges of the masked core (I don't if it is doable or my masks are not adapted).

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: Masked stretch and big star appearance
« Reply #6 on: 2019 September 18 10:32:45 »
You can try this:

MaskedStretch and Stars Cores

Greetings,

Enzo.

Many thanx! Looks promising. I attached an image of my Ha star artefact.

Online acmalko

  • Newcomer
  • Posts: 49
    • View Profile
Re: Masked stretch and big star appearance
« Reply #7 on: 2019 September 18 12:07:36 »
I have the same problem (Canon EOS), especially after a SPF deconvolution.
I use the method above.
Actually, I mix MaskedStretch and HistogramTransformation

Thanks for the idea of convolution on the stars :)

Offline drmikevt

  • PixInsight Addict
  • ***
  • Posts: 112
    • View Profile
Re: Masked stretch and big star appearance
« Reply #8 on: 2019 September 18 17:21:50 »
I can't say if its a good idea or not - its just what I sometimes do.  Of course, its going to work better if you're using a refractor as opposed to something with a spider. 

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: Masked stretch and big star appearance
« Reply #9 on: 2019 September 19 04:44:29 »
You can try this:

MaskedStretch and Stars Cores

Greetings,

Enzo.

I tried it but it works only on RGB images, so you can eventually repair the narrowband after recombination in linear. I was combining them in non-linear after the stretch, so it is not adequate for this strategy. I finally decided to try a different stretching strategy not Masked in order to preserve the stars.

Offline John_Gill

  • PixInsight Old Hand
  • ****
  • Posts: 306
    • View Profile
Re: Masked stretch and big star appearance
« Reply #10 on: 2019 September 20 03:52:32 »
Hi,

Have you tried "ArcsinhStretch"  I do a small stretch and then a larger stretch and then finish with a tweak of the HistogramTransform.

space is not black
John
APM 107/700 apo on CGX mount
ZWO Optics - Autoguiding
ZWO1600mm and filters
... when there are no clouds ...

Offline Bobinius

  • Newcomer
  • Posts: 42
    • View Profile
Re: Masked stretch and big star appearance
« Reply #11 on: 2019 September 20 09:09:36 »
Hi,

Have you tried "ArcsinhStretch"  I do a small stretch and then a larger stretch and then finish with a tweak of the HistogramTransform.

space is not black
John

Hi John,

Yes, and it looks very promising! Actually I tried it at first but it was a bit difficult to balance the Ha and the S and O with ArcsinhStretch and I was not please with the nebulosity. Now that you've mentioned it, I gave it another shot and the stars are really good, I managed to balance the channels before combining them using HT.

All best

Bogdan