M94-Please have a go

[Rodd]

37 minutes ago, Xiga said:

As you can see, there are many ways to skin this cat!

And he's thoroughly skinned--nice image.   That look very natural.

[wimvb]

1 minute ago, Rodd said:

The colors I get using BN and CC are nothing like this.  I get a muddy pinkish gray.  I always use an extracted lightness channel for local support when I use TGV denoise--I saw that in a tutorial.  I will try the decon in linera agin followed by HDR in non linear--I don't think I tried that particular combination before.

BN: use the largest preview that will fit between stars as a reference

CC: same preview for background reference, preview that only has the galaxy (including the ring) for white reference. Turn off structure detection.

[Filroden]

Hi Rodd

Late to the game as usual!

As many have said already, you have good data here. I've tried to do as little as possible in processing as the data is so nice. I only used the RGB. My steps in PixInsight were:

• ChannelCombination
• DynamicCrop
• DynamicBackgroundExtraction
• PhotmetricColorCalibration
• A few small iterations of ArcsinhStretch with a MaskedStretch to finish
• CurvesTransformation on both saturation and RGB
• TGVDenoise
• A light SNCR in green
• HistogramTransformation to bring the background down to about 23/23/23

There are many more steps that could have been applied along the way, including extracting an L channel to apply deconvolution, and applying some noise reduction in the linear stage. Likewise, better processors than me could enhance the contrast much better.

I applied a little sharpening (heavily masked to restrict it to the core) and noise reduction in Adobe Lightroom. Overall, I haven't brought out the halo as much as others but I like the colour balance.

 

[Rodd]

2 hours ago, Filroden said:

HistogramTransformation to bring the background down to about 23/23/23

Your background is exceptional, and the stars look wonderful.  No matter how many iterations of curves for saturation I applied to the stars, I could not get the colors that you have gotten in the larger and medium sized stars.  Also--your stars are fully colored--mine are white with colored rims.

Rodd

[Filroden]

2 hours ago, Rodd said:

Your background is exceptional, and the stars look wonderful.  No matter how many iterations of curves for saturation I applied to the stars, I could not get the colors that you have gotten in the larger and medium sized stars.  Also--your stars are fully colored--mine are white with colored rims.

Rodd

I think you mean your background  It really was good RGB. It didn’t need anything else.

I find star colour can best be done in the stretch, not afterwards. I find the ArcsinhStretch with a value of 2 applied 4 or 5 times gets enough colour into the image and then complete the stretch with either a MaskedStretch or HistogramTransformation. ArcsinhStretch on its own bloats the stars too much if used on its own.

[Rodd]

On ‎4‎/‎27‎/‎2018 at 17:51, wimvb said:

It most definitely is not. The core actually has some very nice detail. In PixInsight you can reveal this by using HDRMultiscaleTransform and a 50% mask. Or use LHE at various settings on clones of the image and combine these clones with Pixelmath.

 

On ‎4‎/‎28‎/‎2018 at 05:07, gorann said:

ut not because of the quality of the data (which is excellent btw), just because RGB can be h

 

18 hours ago, Xiga said:

hanks ag

 

On ‎4‎/‎28‎/‎2018 at 09:43, ChrisEll said:

Rodd, one thing I'd say about the data is that it's very clean, I didn't need to do any NR on it and I only did a tiny smidgen at the very end for luck.

Sorry, but I am having trouble with my mouse--I can't highlight what I want to quote (to avoid quoting images and such).  Also, Ken, I am unable to quote from multiple pages, so I missed you (page 3 is only you so far I think).  Anyway--to all, thank you for the input...it is most appreciated.  It will take me some time to go through the steps in the various styles and see what I can make work best for me.  Meanwhile, I have reworked this data....my best effort to date.  There was only one step I did that I don't remember seeing in this thread "Digital Development".  I tried exponential transformation on the outer ring, but could not make it work--though it did work on masks that targeted only the ring--it made them stronger.  I used Digital Development at quite low settings to accentuate the outer ring.  I found that it did that while at the same time reducing the graininess.  The remainder of what I did is pretty standard per the various descriptions above.  I think the background is a bit clipped--but I can live with it for now.  I think allot more work can be done in the core...of which I am not satisfied with at all.  The blue is to purple, and there is a tough of green. But the longer I tinker the more the image degrades, so I better leave well enough alone for this attempt

EDIT:  at normal viewing its ok--but at full resolution it is still -$%#*

EDIT 2:  A second image is a bit more tinkering.  Not sure if its improved.

 

 

[gorann]

You got a lot of star coulour in there (maybe a bit too much but it looks nice) but you have some odd blotchy noice in the shell. To me it seems that these different algoritms that PI run on the images can sometimes do magic but also mess things up. I try to keep it simple and just keept to what I know PS is doing to my image. Like carefully stretching it with curves and increase the contrast locally with high pass filtering. No magic but also no strange surprises.

[gorann]

Since the coulours are so variable, from red to blue, in the different version both here and of M94 images I see posted on the net, and I have no idea who is right, I took the liberty to mix my blue and Wim's red version @wimvb as a compromise. Should even out any oddities and noise from processing. So here is the Swedish average (or PS-PI average) of you data Rodd:

PS. After posting I see it could benefit from some cropping of the dark edges

 

[Rodd]

28 minutes ago, gorann said:

S. After posting I see it could benefit from some cropping of the dark edges

Looks good

[wimvb]

2 hours ago, gorann said:

So here is the Swedish average (or PS-PI average) of you data Rodd

Then it must be truly "lagom". (Sorry guys/gals, there's no translation of this swedish word; "Goldilocks" best describes it.)

[Rodd]

 

Perhaps I have gone too far--but the core appears a bit more detailed.  The rest of the image is mediocre at best.  And I don't really like the palette of the core--but the structures look fairly decent.  And the stars are tighter. Not as nice a Wims delicate spiral ridges with He regions and blue stragglers hinted at....but the best I can do

 

 

[AstroAndy]

Hi

I saw this post a little late, so, for whatever it's worth, here's my effort.

Andy

[Rodd]

31 minutes ago, AstroAndy said:

saw this post a little late, so, for whatever it's worth, here's my effort.

Nicely done Andy.  There is really allot of detail in the central regions.  I need to refisit this with teh C11Edge.

Rodd

[AstroAndy]

Thanks, Rodd, although on yours the outer ring shows up a bit better. There's always the battle between maximizing the data and keeping it "real", so to speak, and seeing so many interpretations of the same data shows the range of diff. processing; well done on that post. - Andy -