Alternative final processing of OSC Rosette

[old_eyes]

I have been playing around with slightly different processes of an OSC image of the Rosette Nebula, and I am interested in your comments and preferences. Which processing do you think does most justice to the object?

Combining data from a number of projects I accumulated 9 hours OSC data from Pier 5 @Roboscopes in Spain (Tak Epsilon 180, ASI 2400MC Pro, Paramount MX unguided).

I passed the data through my usual Pixinsight process - WBPP, DynamicBackgroundExtraction, BlurExterminator, SpectrophotometricColorCalibration, EZ Denoise, Starnet2 separation of stars and nebula.

The star mask I stretched with ArcsinhStretch.

The nebula I processed in two ways. First MaskedStretch, then either using GeneralizedHyperbolicStretch, focusing on enhancing the contrast on the nebula structures, or CurvesTransformation, trying to keep more of the background nebulosity/dust visible. After the stretch a small amount of LocalHistogramEqualization to bring out some of the structures. Then recombining with the stars using PixelMath and the max(star_mask, starless) function

Using the generalized hyperbolic stretch, I produced this:

[A]

Lots of punch, but I have lost some of the outer nebulosity.

Using the curves transformation, which I have had more practice with, I get this:

[B]

A softer rendition with more of the outer nebulosity and dust visible, but not so much contrast in the main nebula structure.

Well, could I combine the best of both?

This image [C], uses the max function in PixelMath to choose the brighter of the two pixels from the original images - max([A], [B]).

This keeps the background nebulosity and dust, but strengthens the contrast and colour in the main nebula. 

The final image uses PixelMath to average [A] and [B] = ([A] + [B])/2

Really interested in which one you prefer, and any comments on what is wrong with my basic Pixinsight process.

[Ozark Bill]

Very nice work! I’m always a fan of seeing as much of the fine dust and details as possible. These objects don’t usually sit by themselves in the middle of dark space like so many pictures show. 

[ollypenrice]

My views:

A is much too contrasty, the faint nebulosity not sufficiently in play, the bright too insistent.

B is the opposite and too flat.

C  is getting there and D is the best.

There is a fundamental issue with the data, though. The bottom left hand corner is dominated by a greenish-brown background. I don't believe this is entirely spurious because I can see evidence of a dusty region there in my own data, partly because the star count is lower. However, I think the initial steps in DBE and colour calibration have exaggerated it considerably. (On the other hand, it may be an imaging 'First' on this target and I'm talking nonsense!) My curiosity is piqued and I'll take another look at all my raw data.

My instant impression on all versions was that there was a lack of small scale texture.  Where did you apply the NR? I would always avoid it on anything but the fainter signal.

Right, I'm off in search of dust in that lower left corner... 

Olly

Update: the brown dust to lower left is certainly present in my Rosettes and I hadn't paid it enough attention. However, I do think it's over-stated in yours, probably because early operations treated it very differently from the rest of the 'background.' Conversely, I've just given my brown dust a boost and like it, so I'm going round in circles!

Edited February 28, 2023 by ollypenrice

[old_eyes]

23 hours ago, ollypenrice said:

My views:

A is much too contrasty, the faint nebulosity not sufficiently in play, the bright too insistent.

B is the opposite and too flat.

C  is getting there and D is the best.

There is a fundamental issue with the data, though. The bottom left hand corner is dominated by a greenish-brown background. I don't believe this is entirely spurious because I can see evidence of a dusty region there in my own data, partly because the star count is lower. However, I think the initial steps in DBE and colour calibration have exaggerated it considerably. (On the other hand, it may be an imaging 'First' on this target and I'm talking nonsense!) My curiosity is piqued and I'll take another look at all my raw data.

My instant impression on all versions was that there was a lack of small scale texture.  Where did you apply the NR? I would always avoid it on anything but the fainter signal.

Right, I'm off in search of dust in that lower left corner... 

Olly

Update: the brown dust to lower left is certainly present in my Rosettes and I hadn't paid it enough attention. However, I do think it's over-stated in yours, probably because early operations treated it very differently from the rest of the 'background.' Conversely, I've just given my brown dust a boost and like it, so I'm going round in circles!

Olly,

The dust is real I am sure. I can see it in the individual subs if they are stretched to buggery (technical term!). Whether the dust is that colour, and whether it has been overboosted by processing are completely separate questions! 

Processing sequence was:  WBPP, DynamicBackgroundExtraction, BlurExterminator, SpectrophotometricColorCalibration, EZ Denoise, Starnet2 separation of stars and nebula. Followed by separate processing of nebula and stars. So a fairly gentle noise reduction after colour calibration and whilst the data was still linear. I generally get as much of the basics done as possible whilst the data is still linear. Feels more logical.

I will re-process omitting NR and see if the brighter areas seem to have more detail. 

When you are happy with your process, please let me see it. It would serve as a good benchmark for me to compare with my own work. As you know, I am still in the foothills of processing mastery.

[old_eyes]

On 28/02/2023 at 10:50, ollypenrice said:

My views:

A is much too contrasty, the faint nebulosity not sufficiently in play, the bright too insistent.

B is the opposite and too flat.

C  is getting there and D is the best.

There is a fundamental issue with the data, though. The bottom left hand corner is dominated by a greenish-brown background. I don't believe this is entirely spurious because I can see evidence of a dusty region there in my own data, partly because the star count is lower. However, I think the initial steps in DBE and colour calibration have exaggerated it considerably. (On the other hand, it may be an imaging 'First' on this target and I'm talking nonsense!) My curiosity is piqued and I'll take another look at all my raw data.

My instant impression on all versions was that there was a lack of small scale texture.  Where did you apply the NR? I would always avoid it on anything but the fainter signal.

Right, I'm off in search of dust in that lower left corner... 

Olly

Update: the brown dust to lower left is certainly present in my Rosettes and I hadn't paid it enough attention. However, I do think it's over-stated in yours, probably because early operations treated it very differently from the rest of the 'background.' Conversely, I've just given my brown dust a boost and like it, so I'm going round in circles!

Olly,

I did no NR until I had got a starless stretched image. Then I applied a mask to the whole of the nebula and used EZ-Denoise on the background. Boosted the stars a bit more than in the previous image and came up with this:

I don't see a huge difference in detail in the nebula, but maybe you think this is a better balanced view? It certainly makes the brown dust bottom left a bit more interesting.

[ollypenrice]

41 minutes ago, old_eyes said:

Olly,

I did no NR until I had got a starless stretched image. Then I applied a mask to the whole of the nebula and used EZ-Denoise on the background. Boosted the stars a bit more than in the previous image and came up with this:

I don't see a huge difference in detail in the nebula, but maybe you think this is a better balanced view? It certainly makes the brown dust bottom left a bit more interesting.

Now I think that's really talking. The only thing I'd be wanting to do if it were mine is ease down the green in the brown, dusty parts. Not globally, though, because green is already low in some of the darker parts.

Olly

[old_eyes]

2 hours ago, ollypenrice said:

Now I think that's really talking. The only thing I'd be wanting to do if it were mine is ease down the green in the brown, dusty parts. Not globally, though, because green is already low in some of the darker parts.

Olly

OK. I extracted a green colour mask, applied it to the image,  and nudged the green down in the unprotected area. If I pushed it too far, it went from brown to distinctly red, losing what I think is the essential nature of this part of the image. So this is a small shift. Better?

[ollypenrice]

Yes. In Photoshop I'd just bring the saturation down in the brown dust - but that's just me. It's a really nice Rosette. (By the way, it sent me back to my recent RASA version and I brought up those interesting chocolate browns to which this image alerted me. It was worth doing.)

Olly