M83 at 2m focal length

[alocky]

Here's 5 hours of 5 minute LRGB subs (24:12:12:12) on M83 from last night. For years I've had a black and white print of M83 that I bought when visiting Siding Springs in the early 80s, and it's interesting to compare the quality of a glass plate and 150" with a 10" newtonian.
Anyway, I'm quite pleased with this, and will now start obsessing about how to get rid of the white cores in the stars, and hope we get a night without smoke so the blue haloes aren't quite so bad!
Processing was drizzle integration, with a very mild decon on the luminance, a bit of stretching, colour calibration using Excalibrator and all mixed up again in Pixinsight.
Camera was a QSIwsg683, Scope was a Meade 10"f4.5 starfinder with an ASA 1.8x Barlow/flattener, all looking quite small compared to the ASA DDM85 they are sitting on.
Comments most welcome, as well as suggestions on how to tame the stars!
Full res version at astrobin:
http://www.astrobin.com/289574/

cheers!
Andrew.

[Maurice]

Thats nice Andrew..
A bit of background colour & noise on my monitor, but very good nevertheless.

cheers
Maurice

[alpal]

Hi Andrew,
looks good.
An ASA DDM85- & an ASA 1.8x Barlow/flattener - you've moved up in the world!
The halos look good & just mean that the star is blue.
If you want to make such stars all blue it's easy in Photoshop.
Fitswork4 might remove the halos but I doubt it.
If you add noise & blur to the dark background in various amounts in the RGB channels then you'll smooth it out.
You could also use a blurred layer mask to get more contrast & sharpness in just the core.

cheers
Allan

[strongmanmike]

Such a great galaxy huh?...Overall looks pretty good to me Andrew, well done :thumbsup:

Just some suggestions (becasue you asked :thumbsup:) while subtle, I can see the decon applied, it imparts a characteristic dottifying of the detail effect and I think the star shrinking/sharpening is all that has happened to the stars :shrug: but they don't look too bad. I would brighten the inner regions of the galaxy a little too so the galaxy doesn't look as flat and dull.

Again a really good result though :thumbsup:

Mike

[DJT]

Looks good Andrew

Did you use masked stretch for this? If so there is a tutorial on star repair for burnt out cores on the PI site.

http://pixinsight.com.ar/en/info/processing-examples/28/maskedstretch-stars-sores.html

Cheers

[alpal]

Hi Andrew,
I had a quick go at your picture -
I hope you don't mind?
When I split the channels I could see how noisy the blue was.
I added noise & blur to the blue background &
I adjusted the bright stars in the blue channel to give a smaller profile.
I have also increased the brightness of the RGB background & worked on the core
to increase it's: brightness, vibrance & contrast with
a blurred layer mask.
I also reduced the red in parts of the background as it was especially stronger in the top left hand corner
when the background brightness was increased.
I also gave it some HLVG to reduce green looking stars.
Do you think it looks better now?

[RickS]

Nice work, Andrew! You've had plenty of suggestions so far so I'll only give you one :) Try a star mask and one of the multiscale tools to do a blur (remove the first few layers) with Chrominance as the target. This will help add a bit of colour to the star cores. You might want to do a bit of a saturation boost afterwards.

Cheers,
Rick.

[alocky]

Cheers Maurice! I see it on my monitor too, but I set the blackpoint high beacause on a smartphone (which sadly is where 99.9% of people will look at it) it's about spot on.


I did ask, and I'm extremely grateful for anybody who takes the time to help me improve! I agree the decon mainly made the stars look a bit smaller, but I figured it meant it was working elsewhere. You're right about the tendency to turn random noise into blobs - the operator is still trying o sharpen things up even when there's nothing there but noise.





Thanks David, I did - thanks for the link!




Definitely an improvement - Thanks for the effort and I always look forward to your very helpful posts in this forum! I can't thank you and others enough for the effort you put in to sharing your expertise!



Thanks Rick. I've always got time for more advice on PI from a true PI Jedi :). I will definitely give that a go!

Thanks again everyone - although it's nice to be congratulated on putting up an image (it's why we do it, right?) when we get polite, objective and constructive feedback like this we can be sure the next one will be even better!
Andrew.

[alpal]

Thanks Andrew.
I'm not the world's best processing guru but I know a few tricks.
When you sharpen a noisy background it makes it worse.
I have found that it's best to split the channels & work on each one independently.
You'll often find that the noise is different in each channel.
Adding noise & then blurring it can really smooth it out!
All of this has to be done with a suitable blurred layer mask
to work only on the area of concern.
I actually picked up a lot of tricks from Ken Crawford.
here:
http://www.imagingdeepsky.com/Tutorials/MSDLB_Stream/MSDLB_Stream.html

I am also impressed by the results people are getting with PixInsight although
I don't have that program.
Also:
I am amazed at the extra detail you can dig out by using Drizzle when stacking.
I'm still trying to twist Mike's arm to get him to stack with drizzle but
he's too strong for me. :)

cheers
Allan

[alocky]

Thanks again for the positive comments - I've also added a few more hours of data, as well as removed the decon from the workflow and tried one of the tricks to remove the white cores - although this has not worked on all of the stars. I also tried unsharp masking instead of decon, and quite like the result. New version at:

http://www.astrobin.com/289574/B/