Jump to content


Photo

M27 Dumbell Nebula

  • Please log in to reply
22 replies to this topic

#1 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 18 June 2013 - 12:54 PM

Posted Image

This was collected over my last two clear days in June. it worked out pretty well. I ended up taking 10 each RGB @ 30 sec and around 25 each RGB @ 240sec. Total integration is around 6.7 hours. Processed in PI very lightly. Steps as follows:

Started by doing HDRComposition to remove the 'tails' from my NABG camera's stars. Not a wonderful job, but pretty good. I think I should've done even shorter exposures... the tails that are there were there in a 30sec exposure!

1. ColorCombination
2. DBE (20 pixel, .75 Weight, 10 samples)
3. BackgroundNeutralization (hard to find a 'dark' spot with no stars in this image)
4. Color Calibration (used same spot for the 'dark' and used the grouping of stars in the upper left with structure detection 'on')
5. HistogramTransformation (basic stretching)
6. HDRMultiscaleTransform (5 and 1, only to the luminance and lightness, de-ringing on)
7. TVGDenoise (100 interations, allowed convergence, NO masking) This process is amazing at noise reduction. I didn't have to do ANY masking and it really did a great job at getting rid of the noise. It's my new favorite.
8. HDRMultiscaleTransform (3 and 1, only to the luminance and lightness, de-ringing on)
9. CurvesTransform (to increase the luminance channel and greatly increase the saturation in the stars.
10. Applied a star mask and did another CurvesTransform to just M27 and the brighter stars. (This completed the 'normal' image look)

Questions/criticisms/compliments are appreciated. The last step was due to guys on #astrophotography telling me it was quite saturated enough! :)

Here is a picture of my original one last year from June:
Posted Image

Thanks for looking,

#2 CounterWeight

CounterWeight

    Cosmos

  • *****
  • Posts: 8237
  • Joined: 05 Oct 2008
  • Loc: Palo alto, CA.

Posted 18 June 2013 - 01:10 PM

Not being a PI post processor I can't comment on the steps, but the new image seems as though the stars are overcooked and the nebula devoid of detail? I can detect the essential 'apple core' in there but seems the rest is overdone to the point it lacks visual definition as seen in your attempt from last year.

#3 Wouter D'hoye

Wouter D'hoye

    Surveyor 1

  • *****
  • Posts: 1519
  • Joined: 27 Jun 2003
  • Loc: Belgium

Posted 18 June 2013 - 01:13 PM

Hi,

Exept for the colors this is a great image. But imo the colors are not quite right. uusally the nebula is Magenta/red and cyan/blue. Yours is much too orange. other than the colors I like your image. Quite some detail and pretty good stars. I think that if you take another look at the colors you'd have a winner.

Wouter.

#4 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 19 June 2013 - 09:40 AM

Thanks guys. The blue is there when I look at the FITS or the JPEG when you zoom in, it's just not inherently 'popping'. I think I'll mask it off again and try to increase the red/blue via curves to make it more in line with average expectations.

Thanks for the help!

#5 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 19 June 2013 - 09:47 AM

I can see the blue and there is definitely more detail to be had than your last version. But I agree with Wouter that the apple core is too orange/yellow in color, at least compared to how it is normally depicted.

#6 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 19 June 2013 - 09:37 PM

Yea, I noticed that too. I think I'll try to re-color balance it too. If anyone wants to take a wack at it, send me a PM or post your Astrobin username and I'll add you to the public pool.

#7 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 20 June 2013 - 07:44 AM

Not sure if I'll get a chance to take a wack at it or not, but my Astrobin name is the same as my name here.

#8 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 20 June 2013 - 08:44 AM

It's in a public pool if you wanna try.

#9 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 20 June 2013 - 08:52 PM

I took a shot at this, but I wasn't having any success. I think the linear data might be needed.

#10 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 21 June 2013 - 04:44 AM

That data is linear? :) it should be unstretched. Pay attention to the notes, I reversed the r and g channels.

#11 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 21 June 2013 - 06:44 AM

Doh. I didn't know how to get the data. I didn't notice that fits files were actually available. I found them now. My bad. And, yeah, I see in the notes the reversed channels.

#12 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 21 June 2013 - 08:43 AM

This is the best I have been able to do with it. Taming the green channel was wicked.

Attached Files



#13 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 22 June 2013 - 09:15 AM

What did you have to do? Looks really good :)

Chris

#14 Wouter D'hoye

Wouter D'hoye

    Surveyor 1

  • *****
  • Posts: 1519
  • Joined: 27 Jun 2003
  • Loc: Belgium

Posted 22 June 2013 - 12:27 PM

Hmm.. it looks better.. But the colors are still off. I took a quick look in PS. I took the following actions:
a: using hue saturation I shifted the whole imaage more to the green. Alos inclreased saturation slightly
b: the adjustment of the colors affected the overall brightness of the image. This was countered with a curves adjustment.

Below my result. Hope this helps you a bit more to sort out this image.

Attached Files



#15 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 22 June 2013 - 05:21 PM

Yeah. Now that is where we want to be. I wasn't able to get what I wanted. I think PixInsight can do what you did in PS. I just didn't think to try that to get the brown to look more red.

My process was very iterative:

1) Color Combine
2) crop
3) Dynamic Background Extraction (DBE)
4) Color calibration
5) Histogram stretch
6) Dynamic Background Extraction (DBE)
7) HDR Multiscale Transform
8) ACDNR noise reduction
9) Histogram stretch
10) Dynamic Background Extraction (DBE)

#16 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 22 June 2013 - 09:55 PM

Wow. Thanks guys :) I learned a lot! made after, I do my work in Pi so your write up was just what I needed. Wouter, thank you as well.

#17 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 29 June 2013 - 09:26 AM

So... was wondering if I'd screwed something up with my filters and decided to check them. I've been imaging LRG for the last few months....

I'd originally set my filter wheel up RGBL and swapped them around the last time I had the camera down. Unfortunately, my laptop had a PS issue and I never changed my software!

ARgh :) I ended up shooting the 'B' channel for my other images last night before the moon came up. Hopefully, I'll have some correct images later.

*SIGH*.

#18 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 29 June 2013 - 09:32 AM

Oh wow. Well that stinks. It would certainly explain the problems taming green I had working with your image.

I'm wondering why you changed away from RGBL? That is a pretty common setup. I have mine as LRGB (after all, that is the way it is often referred to), and found that isn't the "standard" way of setting them up. Still I like them that way.

#19 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 29 June 2013 - 09:40 AM

I'm trying to reconcile what you have by what I see in the frames you have in the public pool. You had me switch R for G and G for R.

After doing that, the frame that is clearly much brighter than the others is the Green channel (so what was originally labels R).

Anyway, looking at it after the fact, I can well believe that what I have labeled as the Green channel is in fact luminosity.

#20 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 29 June 2013 - 10:49 AM

What's labeled "G" is actually R. What's labeled "R" is actually L. There is no Blue channel :).

I think I had a momentary lapse of judgement. I dunno why I changed it.

#21 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 29 June 2013 - 11:43 AM

Madratter,

Here is what I ended up with. This includes some blends of HA and OIII into the RGB channels and I finally got deconvolution to work.

Posted Image

#22 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 7139
  • Joined: 14 Jan 2013

Posted 29 June 2013 - 12:04 PM

The color is certainly way better than what you had.

I'm no expert in deconvolution, having only begun using it myself. But I can see from the bullet hole in the bright star that you need to mask of the middle of it/and or fooled with the ringing parameters to prevent it.

#23 Raginar

Raginar

    Fly Me to the Moon

  • *****
  • Posts: 6138
  • Joined: 19 Oct 2010
  • Loc: Rapid CIty, SD

Posted 29 June 2013 - 12:29 PM

Posted Image

Here's the Narrowband bi-color version. Pretty legit IMHO. I agree completely with your assessment of the RGB.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics