Star Control - The U235 method

[Uranium235]

First of all, im not sure whether this is common practice for control of stars, but its a slightly modified version of the composting technique used for preventing burn out of bright nebulae/galaxy cores - where we are composting not the object, but the stars instead. This will prove useful for objects such as M45, Propus (Jellyfish), Alnitak (HH nebula), M42, Sadr as well as any other object that has a monster star lurking in-shot. Its good for NB as well as Broadband, but especially BB as that is where most star bloat occurs - not the kind of bloating you get from poor focus, but that which you get when you put a bright star on a sensitive CCD chip for any considerable length of time.

Anyway, the test subject will be the Veil nebula and its bright star 52Cyg, but this method can also be used to knock back some of the smaller stars at the same time.

1) Open your image in Ps:

2) Next, develop it with 2 standard curves and a couple of less aggressive ones to get the image to just under the halfway point, and where 52Cyg is still an acceptable size:

3) Make a new copy of the image:

4) Now, futher develop that copy to the point where you have got most of the nebulosity you want . By that point 52Cyg is now out of control:

5) Go back to your short stretched version, CTRL+A (select all), copy and paste this into your long stretched version to create a new layer (CTRL+C copy, CTRL+V paste). Then, create a layer (vector) mask on this new top layer (its the third icon across at the bottom of the layers tab), now copy the long stretch version into the buffer (you need to turn off the top layer for a moment to select it properly), then select the mask by holding ALT while left clicking the empty mask (white box). Paste the long stretch version into the empty mask with CTRL+V. Here is how your layers should look:

6) Making sure you have the layer mask selected, (re-select with alt+left click if you lose it) use Gaussian blur with a setting of 7.5:

7) Now use the curves tool to knock everything right back so you have just selection of the brightest stars and pehaps a bit of the target object:

8) At this point its a good idea to open up a new view of the image so you can see what changes your tweaks have on the final image. Tidy up the mask using a combination of the eraser, and the lasso tool/gaussian blur to clean up around stars that are embedded within the nebula. It is also at this point where you can increase the strength of the mask by applying a positive curve to the mask - but be careful, take it too far and you will end up with an ugly dark halo:

9: Finally, flatten the image to get your final result. From there it can be stretched a bit more, but your large stars will be nice and tight

Hope this proves useful, so what are you waiting for? Get processing!!!   

Rob

[Tinker1947]

Thank you very muchly.....

[Uranium235]

Youre welcome!

"share the knowledge!" No big secrets here when its comes to processing.

[cfpendock]

Youre welcome!

"share the knowledge!" No big secrets here when its comes to processing.

And I am sure that is exactly what most of us benefit from - and even learn from scratch.  For sure I wouldn't have known anything about image processing without all the help and tips (and encouragement) given on this forum.  Another excellent post, thanks very much Rob.

Chris

[martin_h]

Ive just had a quick bash with this method on a Horse Head image and it managed to control Alnitak quite well. 

[martin_h]

[Uranium235]

Ive just had a quick bash with this method on a Horse Head image and it managed to control Alnitak quite well. 

Awesome   Good to see it working for somebody else.

You just need to be extra careful when cutting back to a bright star embedded within a nebula - I just use the eraser as close to the star as I dare, then lasso it, and blur again to soften the edges of the mask.

[ollypenrice]

This is the proper way to do it!  Excellent and helpful post.

At which point I have to confess to my shamelessly cheapskate shortcut (though I use your method for Trapezium/M42, for instance.) I'm afraid I just get to your point with a hard stretch on top of a soft one and then use a partial (say 15%) soft-edged erasor, quite large for the first application, to take off the top layer. I then reduce the size of the eraser between iterations, working in to the core of the star. Tim quite rightly described this use of the earser as butchery!   However, I can usually make it work.  http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-6shTXhn/0/X3/Horse%20HaLRGB-X3.jpg

Next time I promise to do it properly!

Olly

[AndyUK]

Thanks very much Rob !  I've not yet had the chance to try it (work keeps getting in the way!), but Martin's example certainly proves that it works very well.  Roll on tomorrow night when I can get back and try this on my own 52Cyg blob!

[Uranium235]

ollypenrice, on 27 Aug 2014 - 09:16 AM, said:

This is the proper way to do it!  Excellent and helpful post.

At which point I have to confess to my shamelessly cheapskate shortcut (though I use your method for Trapezium/M42, for instance.) I'm afraid I just get to your point with a hard stretch on top of a soft one and then use a partial (say 15%) soft-edged erasor, quite large for the first application, to take off the top layer. I then reduce the size of the eraser between iterations, working in to the core of the star. Tim quite rightly described this use of the earser as butchery!   However, I can usually make it work.  http://ollypenrice.smugmug.com/Other/Best-of-Les-Granges/i-6shTXhn/0/X3/Horse%20HaLRGB-X3.jpg

Next time I promise to do it properly!

Olly

Well controlled Olly! Its well worth doing becuase you can recover detail that would otherwise get lost in the bloatiness of Alnitak. Its also pretty useful for recovering doubles which sometimes merge into one elongated blob once you process an image (elephant trunk springs to mind).

Andy, no prob mate - if you need any pointers give us a shout

[Martin-Devon]

Great post Rob, I shall definitely try this technique out. Thanks also for setting out a fine example of a processing technique, step-by-step, and easy to follow - many folk are going to benefit from this, so thanks for sharing.

Martin

[michaelmorris]

Thanks for posting this - another technique for the arsenal.

[Uranium235]

Thanks for the feedback guys   I hope to see a lot of tight stars this winter!

[Ewan]

Helpful & appreciated Rob.

I find working just on H data my stars resemble small fuzzy blobs not sure if its just because its Ha data as focus is good.

Need to sort it though as i am doing IC5070 at the moment.

[Rudeviewer]

awesome trick there.. I will definatly be trying this. Maybe give this a try on my trifid nebula shot.