Butchered a Synthetic Prawn (FINISHED)

[Atmos]

Rehashing over old data trying to improve upon it, thought I was making some serious headway until after I had virtually finished processing (just added the synthetic RGB stars), zoomed all the way out and realised that after my LRGB Combination I'd left a lot of rings around stars... Artefacts from star removal.

Getting late and I didn't want to start over again tonight so I figured I'd go HARD! So I did a rough global mask and hit the prawn with a reasonably hard green removal. WHAM! (trying to dial down the green rings around stars). Removing the green left a bit more noise (removing cleaner Ha signal, makes sense) so I did a pretty hard noise reduction. KABAM!

If it weren't for all those ringing artefacts I'd actually be quite happy with it... Minus some of the noise reduction. Given that I butchered it at the last minute, I'm happy with the result. Nice a SMOOTH :D

High Res (http://www.astrobin.com/full/256200/0/?real=&mod=)

[Andy01]

Nice colours Colin but overall the image looks very soft to me?

Maybe some sharpening?

[strongmanmike]

You are tenacious Colin, gotta give you that ;)...well, I recon this looks great now and as you say only the residual rings around the stars are a little annoying. It looks like gas so I certainly wouldn't sharpen it any more at all, perhaps a little more contrast :question:..but even that isn't really that necessary to me as I recon a slight softness enhances this neb.

Ok... let's see version 16 :D

Mike

[Atmos]

Luminance wise it is identical to here (http://www.astrobin.com/252166/G/) but with some butchering in the end just for the sake of it :P I made it very smooooth after some serious noise reduction. Right now I am trying to find a decent way for me to replace the stars without leaving any nasty artefacts. I got close last night but I think I may have been a bit aggressive in the star removal.
Probably a bit too much green removal too :P



Was getting a bit late, far too late to start afresh so I figured I may as well just see what disaster I could get myself into :P I think I need more than ~2 hours per channel.

I actually don't mind the softness, if only I can find a good way to get the same softness without losing some of the whispy detail. Maybe not possible, gotta do something with all this cloud though!

I'll be into the Version 16 before the end of the week :P

[mountainjoo]

Colin, you've been very persistent with star work in your images. I have been thinking about ways to put RGB stars into a NB image, and was wondering if you have tried something naive along the following lines (which doesn't use star removal);

1. create star mask using RGB stars
2. apply star mask to narrowband image
3. PixelMath with perhaps some variation of C*RGB + (1-C)*NB with 0<C<1, appled to the NB image

[Atmos]

I think image processing requires life long persistence hehe

I haven't tried something like that yet I don't think, I'll give it a shot tonight. Only takes a few minutes to do a preliminary test :)

[mountainjoo]

Great, I'm interested to see the outcome. I made a slight change to the formula in my post, should be adding the two quantities instead of subtracting one from the other.

[RickS]

That doesn't look too bad, Colin :)



I've found it very tricky to blend NB and RGB stars without ugly artifacts. My solution has been to just blend the colour component of the stars and leave the luminance alone. This is easy to do if you split the image into CIELAB components.

Cheers,
Rick.

[atalas]

A good butcher is rare these days Colin....you've done well dude.

[Placidus]

Hi, Colin,

Strong Mike is right - it doesn't need sharpening. The softness that Andy mentions is because the histogram zero points need clipping. I had a go and the result looks great.

Best,
Mike

[Stevec35]

Regardless of the processing method I think it looks good Colin

Cheers

Steve

[Atmos]

I had a bit of an attempt with what Jerome mentioned, maybe didn't get it right in Pixel Math but what I have found is that replacing the starts is very easy... Replacing them without effecting the area around the star is more difficult.

Last night I figured that it is considerably easier to remove the starts from a stretched rather than while its linear, appears to cause less ringing for whatever reason.

[Nikolas]

It's awesome regardless of its "butchering"
Not a fan of raw prawns, prefer them barbequed and marinated with garlic, olive oil and oregano

[Atmos]

Well, I think I am down to my final version of the Prawn.... Not quite so butchered anymore. After umpteen star masks and various blends I am down to the smallest of star artefacts :)

High Res (http://www.astrobin.com/full/256200/B/)

The synthetic stars are:
Red: Ha
Green: Ha*0.5+SII*0.5
Blue: OIII*0.85+Ha*0.15

[strongmanmike]

This is like watching a movie of your kids growing up :lol:...love it...probably at about the 15yrs old mark now and maturing fast :thumbsup:

Mike

[Atmos]

Haha well if you look through some of the other Prawns history, it did get worse for a while :P
I think I am going to put the Prawn to rest for now... Need more data :P

[atalas]

Tenacious Colin :thumbsup: now you can use the same techniques to remove the dark rings around your eyes.

[Paul Haese]

The last version looks good Colin. I like the contrasting and the colour is pretty nice too.

[Andy01]

Looks really good Colin. :thumbsup:

Still appears soft to my eye but others have expressed their own views on this. Maybe that's just me agreeing to disagree but I know soft when I see it.

Good result and great flouro colors, a testament to your persistance.

Well done ;)

[Atmos]

Got coffee for that haha :thumbsup:



Thanks Paul, it had been mentioned that I needed to bring the black point back and it has made it more contrasty.



Thanks Andy, I'll have a look at running some slight sharpening over it. In the past I have tried sharpening at the end and been picked up on it :P

A couple of hours every couple of nights for a couple of weeks, the most persistence I've put into an image yet :P

[Nikolas]

That's a Prawn I would eat!

[Atmos]

I'm more of a salmon guy myself :P

Here is another version (http://www.astrobin.com/full/256200/C/) just for you Andy, done a slight sharpen. I can see what you mean by it looking soft, definitely shows up when comparing the two at like 400%.

[RickS]

Were you attempting to put some colour in the stars, Colin, or just trying to remove the magenta? If you're just trying to produce clean white stars then that's fairly easily accomplished by processing a separate lum (with stars) and colour (without) as per JP Metsavainio's tone mapping process.

Cheers,
Rick.

[Atmos]

Actually trying to put in some coloured stars, got some nice blues and reds but there are quite a lot of white.

[Andy01]

Nice one Colin, to my eye that looks much sharper, clearer and suits the image. Well done :)

As Marc would say, "One for the cool wall" :D

[gregbradley]

Very nice Colin. I suspect though your monitor is not calibrated as the yellows look lime yellow on my calibrated monitor but on my bluish biased laptop they look fine.

I have carefully checked my monitor and it is displaying daytime images of scenes I am familiar with perfectly hence my comment.

Nice and sharp and great detail. Well done.

Greg.

[Atmos]

:thumbsup:



My laptop isn't calibrated, just have a MacBook Pro 15" that I do editing in all kinds of lighting conditions :P

[Maurice]

Nice result Colin.
V2 looks great!

Cheers
Maurice

[rustigsmed]

looks waaaaay better than the original version you posted some time ago Col :thumbsup: