This years stab at M16. I havnt done tricolour NB before and this has been processed in PI. Its a quick and dirty from a processing point of view and I am interested in hints on how to reduce the magenta around the stars using PI. I know Rick has a script but also looking for other options as well.
Will not be removing the stars any time soon.
Details
GSO RC8 and STL6303e
3 hours each filter, 20 minute subs
Any guidance appreciated. Wouldnt mind some feedback on the palette as well. I used LRGB combine and mapped SHO thereafter tweaked in curves
Change the identifier of each image, by right-clicking its tab and selecting Identifier. Then, change each to ha, sii and oiii.
Then, open PixelMath, untick the singular expression tickbox (I think that's what it's called), and assign sii, ha and oiii to each line. Leave the last line. Expand the bottom section, make sure rescale result is not selected, change the colour space to same as target, then, drag the little triangle icon on to one of your narrowband images. It'll assemble the tricolour image without the dodgy purple stars.
You can then also play around with the equations for each line:
Hi Dave, nicely composed and I actually don't really mind the magenta stars but understand if you do. Good luck with the PI advice, be interested to see what you come up with
What H suggests is basically what I did for my version. I combined with PixelMath in the colour combo I did (much more whacky than his suggestion ). This of course moves you away from the Hubble Palette.
One of the ways you can deal with the Magenta stars is to make a StarMask and then drop the saturation out of the stars. Making a good StarMask can be frustrating at times, but it's worth playing around to see what you can achieve.
What H suggests is basically what I did for my version. I combined with PixelMath in the colour combo I did (much more whacky than his suggestion ). This of course moves you away from the Hubble Palette.
One of the ways you can deal with the Magenta stars is to make a StarMask and then drop the saturation out of the stars. Making a good StarMask can be frustrating at times, but it's worth playing around to see what you can achieve.
Cheers Chris, will be trying all suggestions. A good learning experience.
Quote:
Originally Posted by strongmanmike
Hi Dave, nicely composed and I actually don't really mind the magenta stars but understand if you do. Good luck with the PI advice, be interested to see what you come up with
Mike
Thanks for the comments Mike. To be frank, I quite like the magenta as well but just want to tone them down a bit. There's not a lot of point making up colours for stars in NB I figure.
Quote:
Originally Posted by Octane
Looks beautiful.
Use PixelMath instead of LRGBCombination.
Change the identifier of each image, by right-clicking its tab and selecting Identifier. Then, change each to ha, sii and oiii.
Then, open PixelMath, untick the singular expression tickbox (I think that's what it's called), and assign sii, ha and oiii to each line. Leave the last line. Expand the bottom section, make sure rescale result is not selected, change the colour space to same as target, then, drag the little triangle icon on to one of your narrowband images. It'll assemble the tricolour image without the dodgy purple stars.
You can then also play around with the equations for each line:
That's a fine rendition. Nice colors. Like Mike i like the magenta stars but this is always an aesthetic discussion. Technically a great image
Thanks Scott, much appreciated.
So I tinkered as you do after all the comments and realised that in the initial LRGB combine, the magenta was muted but there. So i just applied a star mask and then played with curves to get a combination I liked but which retained the Magenta stars, just less in yer face.
Really nice image you have there! I actually prefer the overall colour of the first one but I do agree that the stars in the first were a bit over powering.
At any rate, really nicely done
.... So i just applied a star mask and then played with curves to get a combination I liked but which retained the Magenta stars, just less in yer face.
That's the way to do it! Very nice Dave! Like you and Mike I don't mind the magenta stars.
Very nice detail David but the overall magenta cast bothers me. A selective colour layer in photoshop gets rid of it quite nicely.
Cheers
Steve
Thanks Steve. The colourcast had to go. I took a good look at this again today, and chose to balance the color mapping a bit differently after experimenting with a number of techniques.
Pretty happy with the detail I am getting with this setup.
That's the way to do it! Very nice Dave! Like you and Mike I don't mind the magenta stars.
Thanks for the comment. I tried dropping the saturation down on the stars but it was pants so here they stay.
Quote:
Originally Posted by multiweb
Lovely colors. Great object in NB.
Thanks Marc. Slight change in palette though. Its quite different
Quote:
Originally Posted by Atmos
Really nice image you have there! I actually prefer the overall colour of the first one but I do agree that the stars in the first were a bit over powering.
At any rate, really nicely done
Cheers Colin. I find it hard to replicate the colour between repros. My cop out here is that its narrowband and there arnt to many rules.
Magenta stars are a basic consequence of the fact that for most objects the Ha is much brighter than the Oiii or Sii, so the Oiii and Sii needs a bigger stretch and the stars bloat in blue and red.
If you're using PI then the ColorMask script is pretty easy to use... if I say so myself
Excellent work, David. Nice and deep and clean, well composed and processed.
I think you can safely set the dark point for each channel to just shy of the foot of the histogram for that channel. That gives much better contrast without clipping anything important.
Excellent work, David. Nice and deep and clean, well composed and processed.
I think you can safely set the dark point for each channel to just shy of the foot of the histogram for that channel. That gives much better contrast without clipping anything important.
Star colours look great.
Best,
Mike
Thanks Mike for looking and for the feedback. More tinkering then whilst the full moon is about.
I really like the first colouring but not so much the magenta stars. The nebulosity colouring was quite nice and I really like that myself. The new image I am not keen on the colour scheme but think the detail is a little sharper and of course the stars are pretty good for colour. If you can you can use RGB stars which can be used in either lighten mode or screen mode in PS to sort your star colour out. Another image of good detail for GSO too.
but understand if you do. Good luck with the PI advice, be interested to see what you come up with 
). This of course moves you away from the Hubble Palette.
