The Trifid Nebula CDK17 REPRO 2!

gregbradley · #1 22-10-2013, 08:06 PM

I took this some time ago and only just got around to processing it recently as I catch up on some backlogged images.

CDK17, LRGB, Proline 16803, PME mount.

Crop view:
http://upload.pbase.com/gregbradley/...53044809/large regular size
http://upload.pbase.com/gregbradley/...44809/original large size

Widefield view:
http://www.pbase.com/gregbradley/image/152870200/large regular size
http://www.pbase.com/gregbradley/image/152870200 large size

3 hours plus 1 hour of Ha used for the bright stars in the core.
Greg.

Rod771 (Rod) · #2 22-10-2013, 08:29 PM

That's very nice Greg

Very colourful.

Peter Ward · #3 22-10-2013, 08:48 PM

Post processing is very much in the eye of the beholder....

In this case I'm seeing a very high-key image with highlights effectively burnt out.... which is a shame as I suspect the underlying data is all there (and then some ).

A good M20 for sure... but sorry...not one of your best. I think it needs some top end gamma correction.

IanP · #4 22-10-2013, 10:15 PM

Quote:

Originally Posted by Peter Ward

A good M20 for sure... but sorry...not one of your best. I think it needs some top end gamma correction.

Peter, I think gamma correction will also affect this gorgeous blue nebulosity. I'd go for HDRM-transformation in PixInsight with only the "to lightness" box checked. It will look absolutely stunning.

DavidTrap (David) · #5 23-10-2013, 01:07 AM

Sorry Greg - have to agree with Peter & Ian.

DT

Paul Haese · #6 23-10-2013, 09:02 AM

Good colour Greg. Star shapes look good too and the MMOAG with the ST-i certain has had a positive impact. Looks like the flats are working in this image.

alpal · #7 23-10-2013, 09:03 AM

Hi Greg,
It looks like it was good data.
Did you stretch it in 32 bit FITS files from stacks before processing in 16 bit?

(as per my humble advice here:
http://www.iceinspace.com.au/forum/s...d.php?t=113110
)

cheers
Allan

gregbradley · #8 23-10-2013, 11:20 AM

Quote:

Originally Posted by Rod771

That's very nice Greg

Very colourful.

Thanks Rod.

Quote:

Originally Posted by Peter Ward

Post processing is very much in the eye of the beholder....

In this case I'm seeing a very high-key image with highlights effectively burnt out.... which is a shame as I suspect the underlying data is all there (and then some ).

A good M20 for sure... but sorry...not one of your best. I think it needs some top end gamma correction.

Fair enough. I've done a repro using gamma and offset correction plus shadows/highlights tool and its toned down the highlights.

Quote:

Originally Posted by IanP

Peter, I think gamma correction will also affect this gorgeous blue nebulosity. I'd go for HDRM-transformation in PixInsight with only the "to lightness" box checked. It will look absolutely stunning.

Thanks Ian. I looked for that tool and I couldn't find one called that. I looked some more and I see the HDR Wavelets transformation tool has a lightness checkbox. I tried that but the correction looked a bit savage and damaged the image. No doubt there are ways to finesse this tool. I am fairly new to PI but can use some of it. I have some tutorials paid for I need to sit down and watch and go through it some more. There are some useful tools in there for sure.

Quote:

Originally Posted by DavidTrap

Sorry Greg - have to agree with Peter & Ian.

DT

Fair enough. I've done a repro as per above and I've toned down the highlights. I think what happened is in the effort to bring up the extensive faint blue nebulosity its brought up the bright areas too much with it.

Quote:

Originally Posted by Paul Haese

Good colour Greg. Star shapes look good too and the MMOAG with the ST-i certain has had a positive impact. Looks like the flats are working in this image.

Yes flats work well on bright objects just the faint galaxies I have trouble with.

Quote:

Originally Posted by alpal

Hi Greg,
It looks like it was good data.
Did you stretch it in 32 bit FITS files from stacks before processing in 16 bit?

(as per my humble advice here:
http://www.iceinspace.com.au/forum/s...d.php?t=113110
)

cheers
Allan

Yes thanks for that Allan. I am pretty sure. I'd have to check the workflow but the initial processing I always do is using CCDStack and I think default is 32bits but perhaps its 16. It may be something I need to check in CCDstack to default to 32.

Repro posted to the same links.

Thanks for the feedback.

Greg.

RickS (Rick) · #9 23-10-2013, 11:34 AM

Quote:

Originally Posted by gregbradley

I looked for that tool and I couldn't find one called that. I looked some more and I see the HDR Wavelets transformation tool has a lightness checkbox. I tried that but the correction looked a bit savage and damaged the image. No doubt there are ways to finesse this tool..

Greg,

A couple of options for making HDRMT more gentle are to explicitly mask the image before processing with HDRMT (I use PixelMath to combine a luminance mask with a star mask so that stars and dim areas are protected) or to use PixelMath to blend the original image with the processed version.

BTW, I was surprised that the stellar jets aren't more clearly visible in your image. Did you have bad seeing?

Cheers,
Rick.

gregbradley · #10 23-10-2013, 02:30 PM

Quote:

Originally Posted by RickS

Greg,

A couple of options for making HDRMT more gentle are to explicitly mask the image before processing with HDRMT (I use PixelMath to combine a luminance mask with a star mask so that stars and dim areas are protected) or to use PixelMath to blend the original image with the processed version.

BTW, I was surprised that the stellar jets aren't more clearly visible in your image. Did you have bad seeing?

Cheers,
Rick.

Thanks Rick.

It was imaged some time ago. Probably was and why I didn't process it until now.

Greg.

alpal · #11 23-10-2013, 04:31 PM

Greg,

Quote:

Yes thanks for that Allan. I am pretty sure.
I'd have to check the workflow but the initial processing I always do is using CCDStack and I think default is 32bits but perhaps its 16.
It may be something I need to check in CCDstack to default to 32.

Repro posted to the same links.

Thanks for the feedback.

Greg.

Hi Greg,
the repro is much better & I'd be interested to know if it was stretched in 32 bits?
The bright areas are in a narrow range - & so are the dark areas.
It still looks like those areas could be stretched out.
I have found that FITS Liberator does the hard work for you
when used in compression mode & it's free from NASA.
I have used it in compression mode x^1/5 but your image might
work better in x^1/3 or one of many other functions.
It can then be saved as a 16 bit TIFF for processing in Photoshop
or just about any other package.
Also - I don't know whether all the blue on the left hand side is real or not - it could be a partial blue caste.

I wish I had such lovely data to work from.
I've had 5 months of cloudy weather in Melbourne.

cheers
Allan

strongmanmike (Michael) · #12 23-10-2013, 06:26 PM

Nice Tri-ford Greg

Lovely colours...the megenta may be just a taaaaaad iridescent pink and a taaaaaad saturated...I know, I know, coming from me...

Mike

Bassnut (Fred) · #13 23-10-2013, 06:34 PM

nooooo, flat dynamics (over compressed), no 3d, lacking detail, sorry Greg, stop rushing.

gregbradley · #14 23-10-2013, 07:34 PM

Quote:

Originally Posted by alpal

Greg,

Hi Greg,
the repro is much better & I'd be interested to know if it was stretched in 32 bits?
The bright areas are in a narrow range - & so are the dark areas.
It still looks like those areas could be stretched out.
I have found that FITS Liberator does the hard work for you
when used in compression mode & it's free from NASA.
I have used it in compression mode x^1/5 but your image might
work better in x^1/3 or one of many other functions.
It can then be saved as a 16 bit TIFF for processing in Photoshop
or just about any other package.
Also - I don't know whether all the blue on the left hand side is real or not - it could be a partial blue caste.

I wish I had such lovely data to work from.
I've had 5 months of cloudy weather in Melbourne.

cheers
Allan

No that was all 16bit. I checked CCDStack. You can chose to save in 32 bit floating (I presume floating is the largest resolution as there are others like signed - not sure what that means). I am reprocessing the luminance and I think I need to repro it from the ground up with a different approach. As you say there is more depth in the image and the luminance is coming up nicely.

Quote:

Originally Posted by Bassnut

nooooo, flat dynamics (over compressed), no 3d, lacking detail, sorry Greg, stop rushing.

What you don't like it? Is it just too many stars? Yes it needs a redo and perhaps more data. No worries Fred.

Quote:

Originally Posted by strongmanmike

Nice Tri-ford Greg

Lovely colours...the megenta may be just a taaaaaad iridescent pink and a taaaaaad saturated...I know, I know, coming from me...

Mike

Yeah I think this image is off the rails and I start over from scratch..

alpal · #15 23-10-2013, 09:53 PM

Greg,

Quote:

No that was all 16bit. I checked CCDStack. You can chose to save in 32 bit floating (I presume floating is the largest resolution as there are others like signed - not sure what that means). I am reprocessing the luminance and I think I need to repro it from the ground up with a different approach. As you say there is more depth in the image and the luminance is coming up nicely.

Hi Greg,
Good luck with the re-processing - take your time.
You may also want to try one re-process just with RGB -
as the luminance can overpower bright areas & stars unless masked.
The results will be worth it with such good data.

cheers
Allan

multiweb (Marc) · #16 24-10-2013, 07:23 AM

Colors and stars look real good Greg. I was just expecting a lot more definition in the core given your aperture. The two jets are a must. Maybe separate the 6 central stars too. No pressure.

gregbradley · #17 24-10-2013, 07:27 AM

Quote:

Originally Posted by alpal

Greg,

Hi Greg,
Good luck with the re-processing - take your time.
You may also want to try one re-process just with RGB -
as the luminance can overpower bright areas & stars unless masked.
The results will be worth it with such good data.

cheers
Allan

Thanks for the info Allan.

Quote:

Originally Posted by multiweb

Colors and stars look real good Greg. I was just expecting a lot more definition in the core given your aperture. The two jets are a must. Maybe separate the 6 central stars too. No pressure.

Yeah I liked the colour but the details were not so good. Perhaps it was seeing. I am doing another version where the details are coming up much more sharply but those core stars are burnt out still. Needed some short exposures. Luminance were 15 minutes.

Greg.

AG Hybrid (Adrian) · #18 24-10-2013, 06:46 PM

What an absurd and ridiculous picture. Look at what a Greg can achieve with amateur equipment. If they gave Greg the budget that NASA gave to those astronomers to image the most distant galaxy ever, he would probably be able to image the big bang itself and he would resolve it into sub-atomic particles!

gregbradley · #19 24-10-2013, 06:48 PM

Ok this image showed promise from the start but had some problems. I reprocessed this one a few times taking various approaches. A few versions didn't work out well enough.

This one did. Light deconvolution on the RGB and several strength deconvolutions on the luminance and those median combined. Looked pretty sharp and got rid of the out of focus look.

I found some Ha I took a few years ago to control the core stars so that was handy. Its good to keep the data intact.

I am happy with this one and I have absolutely wrung every last bit out of this one I can think of. Without overprocessing it either. It either comes out fairly quickly or its a fail.

The crop view is probably the best as the stars are all round. I think it was a slightly not properly seated adapter that caused minor tilt giving some elongated stars in the right side. Not too bad though and the widefield view is still pleasing.

CDK17, 3 hours (90 30 30 30 LRGB and 60 of Ha (bright stars in core only though). PME, STi guider, MMOAG.

Crop view:
http://upload.pbase.com/gregbradley/...53044809/large regular size
http://upload.pbase.com/gregbradley/...44809/original large size

Widefield view:
http://www.pbase.com/gregbradley/image/152870200/large regular size
http://www.pbase.com/gregbradley/image/152870200 large size

Greg.

multiweb (Marc) · #20 24-10-2013, 06:58 PM

Now you're talking.