trifid nebula Re-processed - Trifid Nebula in Sagittarius ( Messier 20, NGC 6514 ) - new PhotometricColorCalibration tool

[MikeODay]

Re-processed 12th August 2017 using the new PhotometricColorCalibration tool from Pixinsight.

This function seeks to adjust the colour balance of the image by plate solving the image and comparing the colour of the stars in the image with the colour values for these stars as stored in various databases.

( please click / tap on image to see larger / sharper )

.................

Trifid Nebula ( M20, NGC 6514 )

I manged to capture another 60 odd 240sec images in late July to add to the data I captured at the end of June ( Trifid Nebula WIP )

Trifid Nebula in Sagittarius ( Messier 20, NGC 6514 )

( please click / tap on image to see larger and sharper )'

 

and a crop of the main part of the nebula ...

I am quite pleased with how the colour balance turned out - especially the colours of the stars ( my goal has been to get the colours of the stars as close as I can to how they would look with "daylight" whitebalance and no light pollution / sky glow).

-----------

 

"High Dynamic Range" ( HDR ) image of the Trifid Nebula - built from exposures ranging from 1/8 to 240 seconds in duration. 

Image details:

from nova.astrometry.net:
Size: 52.2 x 35.5 arcmins.
Centre: 18h 2 min 30.8 sec, -22deg 57' 37.7''.
Orientation: up is -88.2 East of North ( ie. E^ N> ). 

Telescope:
Orion Optics CT12 Newtonian ( mirror 300mm, fl 1200mm, f4 ).
Corrector: ASA 2" Coma Corrector Quattro 1.175x.
Effective Focal Length / Aperture : 1410mm f4.7. 

Mount:
Skywatcher AZ Eq6 GT.

Guiding:
TSOAG9 Off-Axis-Guider, Starlight Xpress Lodestar X2, PHD2 . 

Camera:
Nikon D5300 (unmodified) (sensor 23.5 x 15.6mm, 6016x4016 3.9um pixels).

Location:
Blue Mountains, Australia
Moderate light pollution ( pale green zone on darksitefinder.com map ). 

Capture:
12 sets of sub-images with exposure duration for each set doubling ( 1/8s to 240s ) all at ISO800. 

Processing:
Calibration: master bias, master flat and no darks.
Integration in 12 sets.
105 x 240sec main image.
5 each for exposures 1/8 to 120sec - to caputure highlights.

HDR combination using Pixinsight's PixelMath function.

[MikeODay]

I just noticed a small "horse head" and "scream mask" in the image ...

 

Mmmm, perhaps it is more of a "dog's head"

 

[MarsG76]

Nice trifid, I like the glowing effect... 

[Allinthehead]

Another stunner Mike. I'm extremely envious of your skies.

[MikeODay]

23 hours ago, MarsG76 said:

Nice trifid, I like the glowing effect... 

 

16 hours ago, Allinthehead said:

Another stunner Mike. I'm extremely envious of your skies.

Thank guys, much appreciated.

[MikeODay]

Re-processed 12th August 2017 using the new PhotometricColorCalibration tool from Pixinsight.

This function seeks to adjust the colour balance of the image by plate solving the image and comparing the colour of the stars in the image with the colour values for these stars as stored in various databases.

I also went back and re-removed the light pollution from each set of integrated images using a more aggressive approach that sacrificed some faint stars in order to more fully remove the light pollution from the shadows in the image ( this LP was contributing to the colour cast in the faint areas of the images above ).

Tifid Nebula in Sagittarius ( Messier 20, NGC 6514 )

( please click / tap on image to see larger )

 

I am much prefer the colours in this version  ( no matter what I did before I could not get rid of the slight green tinge without causing other unpleasant issues )

[MikeODay]

Another few crops ...

 

 

I love the range of star colours in this part of the sky ...

 

 

 

[Rodd]

Nice image Mike.  I really like this target but its quite low for me and mired in the dregs of a nearby city.  I agree the star colors look good.  I have a question regarding the brightest regions of the nebula--they are whitish.  I have not used the new PI tool yet (don't even know where it is).  I don't recall your earlier version--I am wondering if the whitish regions (usually indicative of a very aggressive histogram) are due to a setting in the new tool.  How tweakable is it?

Rodd

[swag72]

Interesting Mike - I just tried that colour calibration thingie in PI...... I ended up with THE most awful green stars! Won't be using that again  

[Rodd]

6 minutes ago, swag72 said:

Interesting Mike - I just tried that colour calibration thingie in PI...... I ended up with THE most awful green stars! Won't be using that again  

Are there a bunch of settings options?  PI tools have many setting options usually and I have found that with the correct settings, you get perfect results, but with incorrect settings its rubbish.  With little documentation (never available when I need it) its trial and error.  The smallest variation in a setting can really matter.

Rodd

[swag72]

A bunch of settings? In PI? ..... If I said no would you believe me?  

Heartening to see in my experimentation its still as terrible as always  what would I have done if it had worked first time? I may have been a convert

[Rodd]

Just now, swag72 said:

A bunch of settings? In PI? ..... If I said no would you believe me?  

Heartening to see in my experimentation its still as terrible as always  what would I have done if it had worked first time? I may have been a convert

I have not tried it--wouldn't know where to begin.  But it must be able to work....somehow.  Vicent Peris is a pretty knowledgeable processor (computer) and a very good processor (images).  I took his seminar and he knows his stuff.  Believe me, he does not publish images with green stars.  Still may not be worth the trouble though.  You seem to have great success without it.

Rodd

[swag72]

I'm not for one moment suggesting the issue is PI....  Its just not for me and stupidly evey so often I think I'll have a dabble.... It always ends in abject misery

[MikeODay]

13 minutes ago, swag72 said:

A bunch of settings? In PI? ..... If I said no would you believe me?  

Heartening to see in my experimentation its still as terrible as always  what would I have done if it had worked first time? I may have been a convert

 

5 minutes ago, Rodd said:

I have not tried it--wouldn't know where to begin.  But it must be able to work....somehow.  Vicent Peris is a pretty knowledgeable processor (computer) and a very good processor (images).  I took his seminar and he knows his stuff.  Believe me, he does not publish images with green stars.  Still may not be worth the trouble though.  You seem to have great success without it.

Rodd

 

For what it is worth, here are the settings I used ...

[Rodd]

Just now, swag72 said:

I'm not for one moment suggesting the issue is PI....  Its just not for me and stupidly evey so often I think I'll have a dabble.... It always ends in abject misery

Yeah....its definitely a full time endeavor, at least until you master everything, but that takes a full on commitment.  But they all do. I am 100 convinced that there are tools in PS that would benefit my images....but to be able to use them with surgical preciseness, I need to master the program.  Learning 1 was bad enough for me.

Rodd 

[Rodd]

3 minutes ago, mike005 said:

 

 

For what it is worth, here are the settings I used ...

that's worse than I thought!
Rodd

[MikeODay]

2 hours ago, Rodd said:

Nice image Mike.  I really like this target but its quite low for me and mired in the dregs of a nearby city.  I agree the star colors look good.  I have a question regarding the brightest regions of the nebula--they are whitish.  I have not used the new PI tool yet (don't even know where it is).  I don't recall your earlier version--I am wondering if the whitish regions (usually indicative of a very aggressive histogram) are due to a setting in the new tool.  How tweakable is it?

Rodd

Thanks Rodd

The bright areas are nothing to do with the new tool.  This is an HDR image with detail in the shadows not normally seen in images of the Trifid nebula.  In order to bring out that detail I have used a very strong stretch and it is this that has pushed the bright the areas of the nebula.  Basically, the nebula is in fact a great deal brighter than the weak stars and background of this image and I wanted to show that while aslo retaining contrast in the mid-tones.  With at 32bit brightness range something had to be bright

 

[Rodd]

1 minute ago, mike005 said:

Thanks Rodd

The bright areas are nothing to do with the new tool.  This is an HDR image with detail in the shadows not normally seen in images of the Trifid nebula.  In order to bring out that detail I have used a very strong stretch and it is this that has pushed the bright the areas of the nebula.  Basically, the nebula is in fact a great deal brighter than the weak stars and background of this image and I wanted to show that while aslo retaining contrast in the mid-tones.  With at 32bit brightness range something had to be bright

 

If you mask off the rest of the image you can histogram or curve down the bright areas a bit--may reveal some structure and color in there (might have to tone down the saturation too).

Rodd

[MikeODay]

2 minutes ago, Rodd said:

that's worse than I thought!
Rodd

Well ... I have tweaked a couple after reading the help/hints but most of them are the default values.  

[MikeODay]

46 minutes ago, Rodd said:

If you mask off the rest of the image you can histogram or curve down the bright areas a bit--may reveal some structure and color in there (might have to tone down the saturation too).

Rodd

 I guess I was after the colours in the nebula and stars and not so much the detail of the nebula; although the detail is there, it is just in sublte shades and tonal variations ...

I am however quite pleased with how the stars turned out; I love the pale blues and yellows in some of them ...

[Rodd]

Its a fine image

Rodd

[MikeODay]

Cheers Rodd

[gnomus]

On 12/08/2017 at 13:41, swag72 said:

Interesting Mike - I just tried that colour calibration thingie in PI...... I ended up with THE most awful green stars! Won't be using that again  

Its early days.  I got a very green result when I tried to apply the tool to an already stretched image, as opposed to the linear version.  Otherwise, it seems so far to have been fairly reliable.  But ... as I said ... early days.

[MikeODay]

On 15/08/2017 at 04:37, gnomus said:

Its early days.  I got a very green result when I tried to apply the tool to an already stretched image, as opposed to the linear version.  Otherwise, it seems so far to have been fairly reliable.  But ... as I said ... early days.

I have tried the Pixinsight PhotometricColorCalibration tool on two images now and like you I got better results with the linear version before stretching.  

Another observation is that trying to remove light pollution with a DBE base process before using PPC produced better results than trying to apply PPE to a heavily polluted object.

Below is the second example ...                                                                                                  

  <-  Histogram based colour balance process  <-  PhotometricColorCalibration tool

[MikeODay]

One of the earliest photographs of the Trifid Nebula from 1899 ...

 

 

From " Photographs of Nebulae and Clusters, made with The Crossley Reflector" by James Edward Keeler 

The book can be downloaded for free from:   http://www.gutenberg.org/ebooks/36470