Seagull Nebula - PI Experimentation

[gnomus]

Last spring, I posted a Seagull Nebula that was captured and processed at Les Granges - https://stargazerslounge.com/topic/265066-seagull-nebula/#comment-2902769.  

I've been playing with PixInsight more - partly as a result of Mr Keller's book.  I decided I would try the NBRGBCombination Script in PixInsight for adding the Ha and OIII to the RGB image.  The resulting image was interesting.  The stars were horrible, but there was something I liked about the nebula itself.  So I removed the stars using Straton (I had to split the RGB image into separate channels to do that - Straton doesn't seem to work on colour).  There was some tidying up to do after star removal and I did some tweaking to the nebula colour and contrast after recombination.  I then added in just the RGB stars using Photoshop's blend mode 'Screen'.  There were the usual bits and pieces afterwards, but nothing substantial.  The result was quite different to the earlier processing.

• RGB 4 x 400 secs each (yes really - Olly was showing me what could be done in just one night with his dual Tak rig)
• Ha 12 x 1200 secs
• OIII 8 x 1200 secs
• Dual FSQ 106 and full frame Atik & SBIG cameras 
• Mesu

 

[Barry-Wilson]

I really like the effect here on the nebula - it gives a strong impression of tenuous (diaphonus diaphanous . . . well spotted Olly ) gasses.  Less sledgehammer and more feather (not that your processing is akin to a sledgehammer Steve).

You may know that you can replace stars by creating a star mask (needs to be fairly accurate across the range of star sizes and including their halos) to expose the stars and protect eveything else (often an idea to use HT to stretch the mask for better protection).  Then using PixelMath selecting the image with the star colour you want drag the process icon to the protected image.  This will replace the stars.

[ollypenrice]

I like it. As Barry implies, it is more diaphanous (it's with an 'a' Barry, but I had to look it up!!) than the version I did with this data. I find it remarkable that Steve has been able to remove and replace the stars and yet keep them looking so natural. Quite honestly they look more natural, at this resolution at least, than my 'natural' ones...

Barry's images have as natural a look as anyone could ask for, which I think is unusual in PI processing, and we now have Steve achieving this super-natural look as well. I feel like a man on the run...

Olly

Oh, the other camera is not an SBIG, it's a Starlight Xpress SXVH36. I do try to look after our suppliers!

[swag72]

This is looking very pleasing...... I tried a NB of that last year and it never really saw the light of day

@ollypenrice - Have no fear and stop running Olly - there's still some PS folks about  

[gnomus]

Thanks all:

1 hour ago, Barry-Wilson said:

.... Less sledgehammer and more feather (not that your processing is akin to a sledgehammer Steve).

...

Don't worry Barry I know you don't think my processing is sledgehammer-like .... it's more:

 

1 hour ago, Barry-Wilson said:

....

You may know that you can replace stars by creating a star mask (needs to be fairly accurate across the range of star sizes and including their halos) to expose the stars and protect eveything else (often an idea to use HT to stretch the mask for better protection).  Then using PixelMath selecting the image with the star colour you want drag the process icon to the protected image.  This will replace the stars.

 

14 minutes ago, swag72 said:

......

@ollypenrice - Have no fear and stop running Olly - there's still some PS folks about  

Fear not - I used the NBRGB combo script but most of this was done in PS.  That Star Mask thing sounds kinda scary.  I still can't do without my PS comfort blanket.

 

[ollypenrice]

1 hour ago, gnomus said:

Thanks all:

Don't worry Barry I know you don't think my processing is sledgehammer-like .... it's more:

 

 

Fear not - I used the NBRGB combo script but most of this was done in PS.  That Star Mask thing sounds kinda scary.  I still can't do without my PS comfort blanket.

 

CLONE STAMP!!!!!!!

Olly

[PatrickGilliland]

That has worked very well Steve - stars are covered well above, I like the subtle toanl extension it has provided to the nebula, very nice

Paddy

[gnomus]

Thank you all for your encouragement on this one.  I got an email today letting me know it has been awarded EAPOD for 8 February 2017.

[johnrt]

10 hours ago, gnomus said:

Thank you all for your encouragement on this one.  I got an email today letting me know it has been awarded EAPOD for 8 February 2017.

Just noticed this had made an EAPOD, very well deserved.

[gnomus]

7 hours ago, johnrt said:

Just noticed this had made an EAPOD, very well deserved.

That's very kind of you John.

[wimvb]

18 hours ago, gnomus said:

Thank you all for your encouragement on this one.  I got an email today letting me know it has been awarded EAPOD for 8 February 2017.

Congratulations.

Eapod? Where can we admire that?

Ok, got it

[gnomus]

8 hours ago, wimvb said:

Congratulations.

Eapod? Where can we admire that?

Ok, got it

http://eapod.eu/8-february-2017-the-seagull-nebula/

[Barry-Wilson]

Well done Steve for this recognition of your super image.  Keep up the PI processing I say (stands back in readiness for a tirade ).

[steppenwolf]

Very nice image indeed, Steve with some great processing.

[gnomus]

Thank you Steve and Barry.

[RichLD]

Very nice image indeed and congratulations on the EAPOD. Also many thanks for the transpicuous explanation of how you attained such diaphanousness, your processing technique is truly idiosyncratic...

All said in my best Leonard Sachs voice 

Seriously good image sir!

[gnomus]

13 minutes ago, RichLD said:

Very nice image indeed and congratulations on the EAPOD. Also many thanks for the transpicuous explanation of how you attained such diaphanousness, your processing technique is truly idiosyncratic...

All said in my best Leonard Sachs voice 

Seriously good image sir!

Rich by name, linguistically rich, by nature.  

I am gratified by your effusive acclamation.

Steve

[Davey-T]

Congrats on well deserved EAPOD Steve, fame at last 

Mrs Gnomus will be wanting her own one 

Dave

[peter shah]

Nicely done, Love this version over the previous. It would be interesting to understand workflow adding the RGB star layer. You have managed to add them seamlessly,  a superb final rendition.

[gnomus]

Thank you Peter.  I described the essence of the process in the post.  I'd be happy to answer any specific questions you might have from my summary.

Steve

[peter shah]

27 minutes ago, gnomus said:

Thank you Peter.  I described the essence of the process in the post.  I'd be happy to answer any specific questions you might have from my summary.

Steve

I can see you have retained a beautiful star colour, it was more of the balance between the star layer and background. Did you have to use the opacity slider at all to get a smooth transition and did it affect the colour saturation in the stars much?

[gnomus]

11 minutes ago, peter shah said:

I can see you have retained a beautiful star colour, it was more of the balance between the star layer and background. Did you have to use the opacity slider at all to get a smooth transition and did it affect the colour saturation in the stars much?

Thank you.  The star colour is from just the RGB, which I tried not to overstretch.  I pasted this RGB image over the top of my star-free image and set blend mode to 'Screen'.  There was no reduction of opacity - I'm not sure this would have worked.  (Wouldn't the stars end up a bit 'see-through' if you reduced opacity?)

[peter shah]

Perfect thank you for that explanation . This very same method I have used but with mixed results. I think its quite a fine balance, the star layer process has to be a good match with the background to get a smooth transition, a testament to how good your processing is.

[gnomus]

1 hour ago, peter shah said:

Perfect thank you for that explanation . This very same method I have used but with mixed results. I think its quite a fine balance, the star layer process has to be a good match with the background to get a smooth transition, a testament to how good your processing is.

That is very kind of you Peter, but my processing is a lot more haphazard than you suggest.  Sometimes, it works, but, of course, I try to keep the absolute disasters hidden.

I cannot recall every detail of what I did here, but I have a vague memory of thinking that if the RGB background was 'blacker' than the Nebula only image, then blend mode 'Lighten' would only allow the stars through.  I was going to use Lighten, but then thought I'd double check how folks here recommended adding RGB stars.  Screen seemed to be the recommendation so I tried that.