M78-HaSLRGB

[Rodd]

This target has been my white whale--I tried shooting it 2-3 years ago when I started this venture, but the weather and Moon (and assorted gremlins) conspired to make me miss it.  This attempt was started back in August very late in the night, when Orion just popped over the horizon.  I really, really wanted to finish this one this season, my last before changing scopes/cameras/spacings/locations etc.  Now that I have, I feel a bit like the little dog who caught the school bus, or the coon hunter out with his blood hound night after night, hunting the same racoon for years, never able to get him, and when he does, he found that life had lost some of its meaning (that was a story I read somewhere--but the source eludes me)  OK--perhaps exaggerated a bit.  But only a bit.

I do not know why the red came out so magenta-ish.  I added Ha to redden it up a bit, but the straight RGB looked peculiar--purple-ish almost.  For a second I thought I had integrated the wrong stacks, or had called the red the blue.  I even thought I was working with narrow band for a minute, not the SHO palette but the other one (The French team one).  But none of these situations were the case.  I managed to process through it for the most part, thanks to the Ha.

FSQ 106 with .6x reducer and ASI 1600.  The SL in the name stands for super Luminance (1,317 subs-the LRGB subs)

Red: 260 1 min

Green: 267 1 min

Blue: 205 1min

Ha: 50 5min

SL: 1,317 (530 Luminance and the rest the RGB subs)

I think the blacks are clipped, even though the histogram says they are not.  Not what I expected at all as far as palette.  The only real color present was a strange red.  This was the best I could do.  I am glad its over...on to something new.  I will no doubt be processing this data for some time to come (using Ha, not using Ha, straight Lum, super lum, drizzle, non-drizzle, etc).  

Rodd

 

[Rodd]

Here's an alternate version--cropped off the edge artifacts on the bottom.  Slightly different approach.

 

 

[Ceph]

Nice shot Rod, I think I rember that you uses Pixinsight, If you dont know this, there are a magenta reduction script available here
https://pixinsight.com/forum/index.php?topic=7377.0
Not only stars, It works for a full image as well. Could be worth a try?

[Rodd]

4 hours ago, Ceph said:

Nice shot Rod, I think I rember that you uses Pixinsight, If you dont know this, there are a magenta reduction script available here
https://pixinsight.com/forum/index.php?topic=7377.0
Not only stars, It works for a full image as well. Could be worth a try?

That script is typically for narrowband.  There really shouldn't be any magenta in broadband. I might try it if I can't find the source of the problem.  But I have bigger problems....I just woke up and looked at this image and realized the gremlins must have infested my mind.  Wow.  Beyond the pail!

Rodd

[wimvb]

Another excellent image, Rodd. This must have been a nightmare flattening the background. As for the magenta, how are the colours of the HaRGB? Sometimes it's the LRGB combination that causes colour shift.

To correct, you can try adjusting the hue colour curve in CurvesTransformation.

[carastro]

Excellent Rodd, I don't know why you are so self critical most of us would love to get something so beautiful. 

I had been trying to get this one too for several years but was never in a dark location when it was up.  Finally managed it last year, and as you say the Ha really helps with the final result.  

Carole 

[wimvb]

Putting my money where my mouth is:

[Rodd]

6 minutes ago, wimvb said:

Another excellent image, Rodd. This must have been a nightmare flattening the background. As for the magenta, how are the colours of the HaRGB? Sometimes it's the LRGB combination that causes colour shift.

To correct, you can try adjusting the hue colour curve in CurvesTransformation.

Thanks, Wim.  You are too kind (or the gremlins are contagious and have affected YOUR mind!). The only color that was off was the color of the loop--the RGB combined image looked normal as far as colors, except the loop looked strange.  I went back to my subs and noticed that the blue subs had quite a bit of signal in the loop.  I thought maybe I had called red subs blue--but the subs had a B suffix, which is attached in Maxim when blue subs are taken.  So they are in fact blue subs.  The blue tends to make the red look purpentish (that's purple-magenta-ish).  Anyway--I am starting from scratch (post calibration).  Currently using SFS to see if any really bad subs got included (with over 1,300 subs, it takes a while).  This image will be another marathon--like NGC 1333, which ended up coming out ok (only took me a year or so).  The other thing I have to do is insert the Ha better--I can insert Ha into galaxies no problem--but emission nebulae are another story.

Rodd 

[Rodd]

1 minute ago, carastro said:

Excellent Rodd, I don't know why you are so self critical most of us would love to get something so beautiful. 

I had been trying to get this one too for several years but was never in a dark location when it was up.  Finally managed it last year, and as you say the Ha really helps with the final result.  

Carole 

Thanks Carole--the strangest thing happens to me...when I post an image I think it looks great (or at least postable), and am happy.  The next morning I look at the screen and I can't believe what I see.  It is like "what was I thinking!"  Gradually through the day the image starts looking better.  This does not happen with printed images.  So maybe its the screen/light frequency/eye/brain thing (?).  Either way....I mean do disrespect.  

Rodd

[Rodd]

7 minutes ago, wimvb said:

Putting my money where my mouth is:

Nice  Wim.  Still-I have never seen red in an image look the way it did when I combined the RGB.  I guess I need to learn the PI tools a bit better.  I feel like the acolyte circling the temple again and again....at some point a hidden door will become visible and I will be granted access to the next level.  

Rodd

[wimvb]

1 hour ago, Rodd said:

I feel like the acolyte circling the temple again and again....at some point a hidden door will become visible and I will be granted access to the next level

Too many computer games? ?

1 hour ago, Rodd said:

Still-I have never seen red in an image look the way it did when I combined the RGB

If this nebula contains more Hb than usual, it will show signal in the blue channel, and come out magenta rather than red.

[Rodd]

50 minutes ago, wimvb said:

his nebula contains more Hb than usual

Hmm.....Then it must, because the loop is distinct in my blue subs.  I will post the blue stack once PI is done integrating....I eliminated 80-90 really bad subs and am integrating them now.

Rodd

[Rodd]

1 hour ago, wimvb said:

If this nebula contains more Hb than usual, it will show signal in the blue channel, and come out magenta rather than red.

Here's the blue stack (201 1 min).  STF stretch only (and crop and DBE).  Pretty distinct loop.

[Rodd]

Well--here is a reprocess using the NBLRGB script to add Ha to the red channel.  I guess I was doing things about right since I didn't see much of a difference.  I tried to be more careful with processing in general. Nope--too much noise control for full resolution viewing.  I can see this is going to be tough.  

Edit:  Nope, too much noise control for full resolution viewing.  Darn

[Rodd]

This is as good as I can muster.  I like parts of it--and other parts not so much.  M78 itself is one part I like.  It was very difficult to prevent the clipping of the dark structures.   I would like to drop the brightness overall, but I do not want to darken the dark structures.

 

The best part of the image

 

 

 

[Rodd]

To wrap things up--here is an LRGB version-no Ha.  Still pushed a bit to much--but getting there

[Rodd]

Well--the end of a painful road.  In order to get the nebula where I want it, I had to push it a bit further than the data can take.  I have been trying to mitigate this exceedance.  This one does have Ha in the red channel.  I much prefer the full resolution viewing on an image with about half as much processing--but the normal viewing of that image is dull.  So this one is better at normal viewing--but a bit too much for full res.  A trade off it seems.  I suppose in the end processing for normal viewing makes more sense.  Maybe some day I will be able to shoot this target from a dark site.  The crop is a straight crop--no additional processing.

Full image

Crop

[Rodd]

OK--I need help.  Here is the image with 1/2 the processing.  It seems dull at first glance, but maybe I am asking to much of the data.  It certainly would not be the first time Should this image be the final version?  or the more processed, more striking image?  Opinions needed.  Thanks.  

Rodd

[wimvb]

I like the previous, bolder version better. The contrast between the red, dark and blue is what this target is about, imo.

[Rodd]

5 hours ago, wimvb said:

I like the previous, bolder version better. The contrast between the red, dark and blue is what this target is about, imo.

Thanks Wim.  I agree with the premise, but I am not sure the data can handle it.  Each time I process this image I get to a point where there is no visible noise, no artifacts and it looks very clear--even at full resolution.  But a bit dull.  All efforts at lifting the signal to achieve the more dramatic result end up causing the dimmer areas of the image to look "chunky" or "curdled" (for the lack of better descriptors), or inhomogeneous.  Thin gas is supposed to have a degree of homogeneity between structures, at least at the pixel and small pixel cluster scale.  In this image this becomes visible at full resolution viewing.  Darkening is a way to hide the problem--but clipping of the dark structures, and the small amount of visible background space, results if I go much further (even now perhaps).

Rodd

[wimvb]

Have you tried @Barry-Wilson's routine for stretching?

https://barrywilson.smugmug.com/PixInsight-Tutorials/Boosting-star-colour-Repaired-HSV-S

 

[Rodd]

10 minutes ago, wimvb said:

Have you tried @Barry-Wilson's routine for stretching?

https://barrywilson.smugmug.com/PixInsight-Tutorials/Boosting-star-colour-Repaired-HSV-S

 

Page not found.  I have started to remove some of the violet cast.  Here's the latest.  I think it's improved BUT--I seem to have lost some of the reddish hues in the dust lanes of M78.  Darn--just noticed that.   I used the SHO-AIP script for this.  I noticed a strange difference when I used HDR compression in the core of M78--it blew it apart.  I had to extract the L, compress that, then reinsert it, which worked.  Full resolution getting easier on the eye.  Question--the next 2 nights will be clear (and extremely cold).  I could capture 4-8 more hours on this.  Which channels do you think would be the best.  Or maybe I don't need more data.

 

[Barry-Wilson]

Hi Rodd

The page is here, https://barrywilson.smugmug.com/PixInsight-Tutorials/Boosting-star-colour-Repaired-HSV-Separation

?

 

[Rodd]

9 minutes ago, Barry-Wilson said:

Hi Rodd

The page is here, https://barrywilson.smugmug.com/PixInsight-Tutorials/Boosting-star-colour-Repaired-HSV-Separation

?

 

Thanks, Barry--one thing I notice is that the repaired star is indeed colored through and through, but seems to me that the cores of star, being so much brighter, should have a lower apparent saturation due to the brighter a light the less color distinction can be made by human eyes.  The repaired star has a flatter appearance--almost like a planet.  Maybe that is just because it is zoomed in?.  I suppose the brightness can be modified.   I guess its worth trying,  Thanks

Rodd

[Barry-Wilson]

It is only a stage in the processing and not the final product remember. The star profile will change when luminance is added.