NGC3576 "natural" colour

[Shiraz]

Hi

All the colour in this one comes from RGB data, colour balanced on foreground stars and with blue pushed a little to partially compensate for the dust extinction. It was composed using the RGB colour data and Ha as luminance, so the stars have essentially undefined outer regions - minimised as much as possible, but still noticeable.

full res:http://www.astrobin.com/full/247005/None/?real=&mod=

Thanks for looking. Regards Ray

[multiweb]

Love it. Top shot Ray. :thumbsup: Always liked this one how one side is orange and the other more reddish. All this is lost in NB.

[RickS]

A lovely image and an interesting contrast to all the narrowband versions we've seen lately. Well done, Ray!

[Atmos]

Really nice rendition Ray :) I suppose one of the benefits of using Ha as your luminance is that you do not become overpowered by stars. Luminance picking up far more stars than Ha.

[Shiraz]

thanks Marc - yes, the actual colour is interesting, but the stars can be overwhelming in LRGB.
Thank yoy Rick!


That's the main reason to do it - the stars are much more subdued but they and the nebulas retain "natural" colour


regards Ray

[astronobob]

Really noice result Ray, mm, interesting, new to me but great to now know Ha is better than Lum for this star preservation issue - they look very nicely defined, pin point as should be - top show & great result :thumbsup:

[gregbradley]

A great rendition. Tons of sharp detail. I love the bright areas of nebula, they look like a rip in the physical universe or a river of lava.

Greg.

[Placidus]

Wonderful contrast in feel between the left and right sides of the image.

Very nice indeed.

[alpal]

Hi Ray,
that's fantastic.
get a print made & hang it on the wall.

cheers
Allan

[Shiraz]

thanks Bob. The technique has it's problems, worst of which that the (bigger) stars in RGB cover up some of the surrounding nebula and you end up with regions around the (smaller) Ha stars where the colour information is that of the star, but the luminance is that of the nebula - leads to odd coloured halos around the stars. I tried a variety of methods to suppress this effect, with fair success, but not total.


thanks Greg - yep, I like the drama in this object as well


thanks M&T, appreciated


Thanks Allan - that is an idea.

regards Ray

[Atmos]

Do you think this issue could be fixed by getting just a small amount of luminance data and blending it in with the Ha?

It would probably cause some stacking artefacts and would be a rejection nightmare but you may even be able to integrate them together just to get some star reduction without having too much of an effect on the nebula regions.

[Shiraz]

that is a good idea Colin. I used heavy deconvolution and morphological erosion on the colour data and a little masked dilation on the Ha stars to get the two sets of stars matched up fairly well, but adding in some luminance may be a better approach - slightly larger stars would not be too much of a problem. thanks for that.

[Atmos]

I have in the past tried integrate vastly different subs and it has always ended up in complete chaos. The difference in the Ha and luminance will help the stars but you WILL get some very bad specklding in the nebula regions.

What you might be better off doing is making a star mask on the Ha data and using PixelMath to incorporate the luminance stars. Off hand, you may need to blur/soften the star mask so as to preserve the light falloff in the luminance data.

Conversely, you could make a star mask in the luminance and drop the Ha data (nebula) into that.

[strongmanmike]

What?...:scared2:

Excellent result Ray...probably overall the best 3576 so far this IIS season :)

Enjoyed the view :thumbsup:

Mike

[Shiraz]

thanks Colin - some more good ideas to try.



yep, deconvolution it was...

thanks Mike - very generous.

regards Ray

[janoskiss]

Amazing. Enjoyed touring that one at 200% zoom. So much to see in this image!

[Shiraz]

thanks very much Steve. Glad you enjoyed it.