Gamma Cygi HaRGB and HOO Comparison

[Rodd]

I wanted to do a broadband image of this nebula but wasn't sure if I should do an LRGB, HaLRGB, HaRGB.  Combining Ha for nebulae is something still new for me.  In the end, I chose an HaRGB composition.  I added a percentage of Ha to the red channel and made a synthetic luminance out of all RGB and Ha subs--that way, Ha would be in the luminance to some degree.  I also chose an Ha sub as the reference when integrating all 411 subs.  The first image is the HaRGB and the second image is a bicolor HOO image with RGB stars.  I know there is a blue halo issue in the HaRGB issue around some stars, and the detail is not what I expected.  I was able to avoid the "pinks" though.  All Data was collected with FAQ with .6x reducer and ASI 1600 camera with Astrodon filters (Series E Gen II True balance and 3um narrowband).  A Hubble Palette image is on its way.

Ha 53 5min

OIII: 74 5min

Red:: 25 60sec

Green : 122 60 sec

Blue: 110 60 sec

SL: 410 (RGB 60 sec, Ha 5min)

HaRGB

 

HOO with RGB Stars

[Adreneline]

Very nice images Rodd.

I like them both but I have a preference for the HOO with RGB stars which in my view reveals more detail.

Thank you for sharing.

Adrian

[Rodd]

26 minutes ago, Adreneline said:

Very nice images Rodd.

I like them both but I have a preference for the HOO with RGB stars which in my view reveals more detail.

Thank you for sharing.

Adrian

Thanks Adrian--and I agree.  I was surpised to see that some prominent details in the HOO image are actually absent altogether from the HaRGB image

Rodd

[Rodd]

Here is an HaHOO image with rgb stars--I think this is better than the HOO image.   The second image is a bit less saturated,  Are the details better in that one?

 

 

 

 

 

 

 

[HunterHarling]

Excellent! I love the HaRGB. 

[Rodd]

2 minutes ago, HunterHarling said:

Excellent! I love the HaRGB. 

Thanks--I am looking forward to the HaSHO

Rodd

[Rodd]

Finally--here is an LRGB composition where the Ha was used in the synthetic luminance, but no Ha was added to the red channel or the Lum channel other than including it in the SL integration.  King of muddy--but possibly more natural?  The blue channel really gave me trouble--the FWHM was way higher and the stars were very much bigger.  Kind of weird, because all other filters were ok.  That is why there are some blue rings that I really tried hard to minimize.  Why would one filter act up like that?

[wimvb]

Rodd, I liked the detail of the HOO with RGB stars in your original post, but the colour of the nebula in the HaRGB of that post. So I downloaded both jpegs, extracted the L (L*a*b mode) from the HOO image and the a and b components from the HaRGB image and combined those to produce this:

[Rodd]

1 hour ago, wimvb said:

Rodd, I liked the detail of the HOO with RGB stars in your original post, but the colour of the nebula in the HaRGB of that post. So I downloaded both jpegs, extracted the L (L*a*b mode) from the HOO image and the a and b components from the HaRGB image and combined those to produce this:

An interesting approach--I will look into that--did not know I could extract all that stuff--how do you extract in CLAB mode--I thought that was just for combination  I have only ever been able to extract LRGB channels.  I do like the color of the bright red areas--but I think I like the colors in my 3rd image posted overall--I like seeing the blue differentiated a bit.  Maybe I do need some good ole fashioned L.  Or SII--the Hubbel palette images of theis region I have seen are pretty good.

Rodd

[wimvb]

40 minutes ago, Rodd said:

how do you extract in CLAB mode

Channelextraction under colorspaces or channelmanagement. The process is the opposite of channelcombination. As in channelcombination, you can choose from different modes.

It would be interesting to see this in the Hubble palette.

[Rodd]

4 hours ago, wimvb said:

Channelextraction under colorspaces or channelmanagement. The process is the opposite of channelcombination. As in channelcombination, you can choose from different modes.

It would be interesting to see this in the Hubble palette.

Thanks--Back to the drawing board--for all the images!

odd

[wimvb]

(oops) I often find that when I rework old images after learning something new, they don't necessarily come out any better. It's just that I understand the "why" better, or accomplish the same thing in a new, and hopefully more efficient way.

Also remember that lightness/luminance affects colour. That's the basis of the colour wheel. It's the reason why the lower right of the image I posted, still has gray in it. Even though the HaRGB has more red.

[Rodd]

2 hours ago, wimvb said:

(oops) I often find that when I rework old images after learning something new, they don't necessarily come out any better. It's just that I understand the "why" better, or accomplish the same thing in a new, and hopefully more efficient way.

Also remember that lightness/luminance affects colour. That's the basis of the colour wheel. It's the reason why the lower right of the image I posted, still has gray in it. Even though the HaRGB has more red.

Well, the broadband image definitely did not meet my expectations.  Maybe real luminance is needed. Or, and possibly more likely, I don't know how to add Ha into RGB data the right way for nebula.  Maybe the STT-8300 would be better (I am getting it backl soon),  The narrowband is nice though, and I am looking forward to adding SII as I got a respectable amount of Ha and OIII, so the HaSHO image should be nice.

Rodd

[Rodd]

On 07/08/2018 at 17:34, wimvb said:

Rodd, I liked the detail of the HOO with RGB stars in your original post, but the colour of the nebula in the HaRGB of that post. So I downloaded both jpegs, extracted the L (L*a*b mode) from the HOO image and the a and b components from the HaRGB image and combined those to produce this:

Wim--you were right.  I did what you said--with some additional tweaks, and I think its the best image that contains RGB I have.  The extracted a and b look like terrible nothings, but when they are combined, it like magic.  thanks for the continued tutoring.  Quite amazing.  Some stars are still bad--the blue subs were giving me a problem ll night.  Not sure why.  Maybe the guide star was too big and while round, the stars were just boated.  Not sure. I am getting SII tonight--I think the HaSHO will be the best.

Rodd

EDIT--I am not so sure my additional tweaks were good after all.  Oh well.  Great process to keep in mind for the future though.   I'm done with broadband for this target--will concentrate of NB.   I would have to recapture the blue, and quite frankly, I am not feeling compelled.

 

[Rodd]

 

On ‎08‎/‎08‎/‎2018 at 04:14, wimvb said:

I often find that when I rework old images after learning something new, they don't necessarily come out any better. It's just that I understand the "why" better, or accomplish the same thing in a new, and hopefully more efficient way.

OK--I had to untweak and tweak again--here is my final version.  thanks again Wim.  Stars are not perfect--but as good as they will get without PS--which now, tahnsk to Dave-T I have--but have yet to climb the mountain

Rodd--2 versions--accentuated blue in the regions that have it (maybe too much)

 

 

[Stub Mandrel]

Nice Rodd.

Also good to see the pink bit at 7 o'clock of Sadr on my own images isn't an artefact!

[Rodd]

1 hour ago, Stub Mandrel said:

Nice Rodd.

Also good to see the pink bit at 7 o'clock of Sadr on my own images isn't an artefact!

Thanks Neil.  I was wondering about  that at first too  but it is a little nebula--or extension

Rodd

[Rodd]

2 hours ago, Stub Mandrel said:

Nice Rodd.

Also good to see the pink bit at 7 o'clock of Sadr on my own images isn't an artefact!

 

On ‎08‎/‎08‎/‎2018 at 04:14, wimvb said:

(oops) I often find that when I rework old images after learning something new, they don't necessarily come out any better. It's just that I understand the "why" better, or accomplish the same thing in a new, and hopefully more efficient way.

Also remember that lightness/luminance affects colour. That's the basis of the colour wheel. It's the reason why the lower right of the image I posted, still has gray in it. Even though the HaRGB has more red.

H ate to do this to you--but by Jove I think I got it--a 50 50 blend of the HaLRGB and Hoo. 

 

[Allinthehead]

Nice Rodd. I like the last one best. If you can sort out those blue stars you'll have a cracker.

[Rodd]

2 hours ago, Allinthehead said:

Nice Rodd. I like the last one best. If you can sort out those blue stars you'll have a cracker.

Thanks Richard--I don't think its to be with this data.  Its almost there.  I plan to do a HaSHO to round up the presentation (HaRGB, HaHOO and HaSHO).  I would have to reshoot the blue channel on a good night.  Maybe--we'll see.  Now that I have PS--maybe I can wave the magic wand and layer the blue halos out.  Don't hold your breath!

Rodd

[gorann]

Very nice Rodd, as usual! Interesting thread.

Like you I also got a bit annoyed with the blue ringed stars so I started thinking about ways to fix them. The problem is as you say that the stars were bigger in the blue channel. My first attempt to fix them involved extracting the blue channel from your last image and then shrink them, but this did not really help and I think it could only work if it is done one a relatively unprocessed version of the blue-filter data (since there were rings around the stars even in the blue data). So then I tried a bit more brutal approach by selecting all the stars, expanding the selection so that i covered the blue rings, and the using curves on the different shannels to lighten the rings up and turn the blue down. Then using layer masks I applied this only to stars in the brighter regions (or there would have been bright rings around the stars in the dark regions). It made it a bit better but more of a symptomatic treatment than a cure. All done in PS of course so I am sure that Wim @wimvbwill tell me that there are much more sofisticated ways of doing this in PI. Also done a bit quick and dirty.

[peter shah]

lovely crisp detail

[wimvb]

8 hours ago, gorann said:

All done in PS of course so I am sure that Wim @wimvbwill tell me that there are much more sofisticated ways of doing this in PI.

There may very well be. But you've solved the problem, so no need to look for a method in PI. As you pointed out, it would be better to fix the problem in the unprocessed blue master.

Excellent rework, anyway.

[gorann]

1 hour ago, wimvb said:

There may very well be. But you've solved the problem, so no need to look for a method in PI. As you pointed out, it would be better to fix the problem in the unprocessed blue master.

Excellent rework, anyway.

Thanks Wim. Looking at it again I think that Rodd really found a nice colour tone. It looks like some very fancy and expensive marble! Rodd, if you are in for a new bathroom maybe you could get the image printed on the tiles?

[Rodd]

10 hours ago, peter shah said:

lovely crisp detail

Thanks Peter!

Rodd