The many colours of Omega Cen

[RickS]

I have been testing an early release of PixInsight 1.8.5 which has some cool new features, one of which is the PhotometricColorCalibration process. PCC uses plate solving and photometric data for stars in the FOV to calculate the colour calibration parameters. You can select from a range of white references.

The attached pics are:
- "Classic" ColorCalibration, stars white ref
- "Classic" ColorCalibration, stars white ref, corrected for galactic dust extinction
- PhotometricColorCalibration, G2V white ref
- PhotometricColorCalibration, G2V white ref, corrected for galactic dust extinction
- PhotometricColorCalibration, average spiral galaxy white ref
- PhotometricColorCalibration, average spiral galaxy white ref, corrected for galactic dust extinction

The data was captured back in 2013 with the Ceravolo 300 @ f/4.9 and Apogee U16M camera from light polluted inner Brisbane. There's approx 2.5 hours of LRGB in 180 and 300 sec subs.

Big versions on Astrobin: http://www.astrobin.com/299315/F/?nc=user

Thanks to Juan Conejero for all the work he has put into 1.8.5 and Vicent Peris for the PCC tool and the model for extinction correction.

Cheers,
Rick.

[multiweb]

That sounds great Rick. When is it coming to release?

[Atmos]

Looks fantastic! D to my eyes looks pretty spot on.

[Placidus]

Such a vexing problem. That there should be six completely different images, all with good scientific justification for their existence, says it is difficult.

Last night, encouraged by Marc Aragnou's slightly wider field image of omega, and dismayed by our omega overflowing the field, we did a tight 5 panel mosaic to get just that little bit of background. There are some bright G0 and G5 stars in the field. That might help a bit.

NASA has an ageing Hubble shot (https://www.nasa.gov/mission_pages/hubble/multimedia/ero/ero_omega_centauri.html) of the core of Omega. They say that most of the stars in the image are sun-like stars. They've chosen to make them yellow-white, rather than white. Subjectively, we like that, but dislike their intentional over-saturation to bring out the smattering of blue stragglers and red giants.

We reckon at making Omega (or giant ellipticals, for that matter) look positively blue is a common mistake which goes against the seniors-card carrying astrophysics of these venerable old objects. We agree with Colin that D looks very pleasing.

Thanks for your timely discussion and examples. It will help us with our processing.

Best,
Mike and Trish

[Andy01]

That's fascinating Rick but a bit confusing for a layman like me - they all look pretty good so which one is supposedly the correct one? :shrug:

[RickS]

Knowing how software development works I try not to ask "are we there yet?" too often :lol: Probably some time in the next couple of months.



Thanks, Colin. Yes, that's a nice colour balance if you prefer the redder end. I think the very blue images of globs that we see aren't very realistic.



Thanks for the well thought out comments, as always M&T! It will be nice to have a scientific, yet flexible, basis for colouring images ;)

[SimmoW]

Wow, that's an impressive tool, even more justifying my embracing the Dark Side of 'your'cult Rick!

Yes, interesting that there are a variety of scientific justifications for different colours. I suppose that what layer masks are for :screwy:

Hurry up and get it released!

[Atmos]

You've got it right there. All globular clusters are red, some are redder than others though. The younger ones (9-7 Gyrs) have sun like stars but they only are not as common as the 10-13 Gyrs that make up the majority of the globular clusters surrounding the Milky Way.

Most of the globular clusters have a upper mass of 0.9 solar masses, a GC like Omega Cent, being one of the older ones, is around 0.8 solar masses. This means that without interstellar extinction it should still be reddish and not the white average neutral that it is generally shown as.

The reason it is generally shown as being white is because 0.8 solar masses becomes the average white because that is the average. Marc's was nicely resolved and perfectly colour balanced, just not scientifically :P

[RickS]

Andy, the first two images (the very blue ones) use the averaged colour of all the stars in the field as a white reference. This is probably justifiable if you have a field that contains a "typical" mix of star types, but for a globular cluster I would say the results are incorrect.

Using a G2V star (same as our sun) as the white reference gives us the two reddest images. I'd consider these images have a high degree of correctness for life forms that grew up on Earth or other planets revolving around a G2V star :) The version with correction for galactic dust extinction shows the colour as it would appear from outside the Milky Way.

Using an average spiral galaxy as a white reference can be justified on the basis that a galaxy contains a nice mix of stars, and not everybody grows up looking (not directly, please!) at a G2V sun. These versions also have a high degree of correctness, IMO, and might appeal to silicon based life forms.



Our cult gets the best stuff, Simon, and PhotometricColorCalibration rolls off the tongue nearly as well as "Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn."

Layer masks are for wusses that can't see numeric L*c*h* values in their head :lol:

I hope it will be out soon but that's up to Juan... I'm really looking forward to large scale pixel rejection. No more painstakingly removing satellite trails :thumbsup:

[gregbradley]

What does Gyrs stand for?
Greg

[Atmos]

Sorry, giga years (billion years) :lol:

[atalas]

Damn....can't acuse you of getting the colour wrong,or right,or wrong,or right.....:confuse3:

[RickS]

Obviously I need to post at least six versions of all my images :D

[multiweb]

You might launch a new trend and make animated gif files from now on.

[RickS]

Here's one I took earlier: https://s-media-cache-ak0.pinimg.com/originals/6c/93/de/6c93de2ae7118c79c79b750462814002.gi f

[Slawomir]

Thank you for sharing colourful fruits of your experimentations Rick :thumbsup:

RGB imaging is something I will have to learn one day, but with my aging cons and rods versions A and F result in the most pleasing sensations in the brain.

Looking forward to experimenting one day with the new version of PI. Hopefully I will have new data by then :prey:

[RickS]

Thanks, Suavi. Yes, you really should try some RGB from a dark site one day.

There's been just enough bad weather around that I haven't been tempted to set up in the back yard. I'd like to get some new data too!

Cheers,
Rick.

[Retrograde]

Very interesting.

To my eye the second last one looks about right (so I'm probably a silicon-based life-form another solar-system) :ship2:

[RickS]

Thanks, Pete. I thought you had an odd shimmering look when viewed with averted vision but I had assumed this was just the Astrofest booze :lol:

[Placidus]

Continuing with my mission to miss the point entirely, one thing I like about all of them is you can see the slightly elliptical shape quite clearly. It seems that is a bit difficult to bring out.

Best,
Mike

[strongmanmike]

If you blend C, D and E....it will be perfect Ricki :D

Yes, the elliptical nature of OC is often mistakenly processed out by people, it is not round, nor does it have a bright concentrated core.

Mike

[Bart]

I like them all. Omega is my most favourite visual object and I love to see images of it.

[lazjen]

Interesting set of pics and topic. I never knew there was so much choice. :)

I guess I'll have to add this to my processing steps as well. Is this calibration performed as the "last step" after all other adjustments?

Also, it looks like you've blown out the core of the cluster, Rick, so you'll have to go back and do it all again. ;)

:thumbsup:

[RickS]

I hope that's not collimation error, Mike :lol:



Ta Mike. I can do that with PixelMath, no problem ;)



Thanks, Bart.



Hi Chris, the colour calibration has to be done with linear data, so it's a very early step in the RGB processing.

I think I've processed that data enough times already. You'll have to take it or leave it :lol:

[SimmoW]

Congrats on the IOTD over at Astrobin Rick! So you'll post each version each day??

[RickS]

Thanks, Simon. I'm going to do a whole year of Omega Cen variations :D

[codemonkey]

Nice work Rick, and congrats on the IOTD. Also thanks for the heads up on 1.8.5, didn't know it was coming. As someone who struggles with getting RGB colour "right" I look forward to playing with PCC... and making more "not quite right" versions :D

[RickS]

Thanks, Lee. I'm a big fan of not-quite-right myself :D

[mikeoday005]

Thanks for sharing your experiment.

This looks like it could be a very useful feature for me as I really struggle with getting 'correct' white balance after I have removed the quite strong light pollution in my area. In the end I usually give up and go with what I think looks most pleasing. It will be good to have a reasonable basis to start from.

[DJT]

Interesting stuff. Looking forward to the new process and playing with old glob data.

Thanks for sharing!

[RickS]

Thanks, Mike. Yes, I think it will give a good starting point to work from for those of us, including me, that are colour challenged :)



No problem, David.

Cheers,
Rick.