Jump to content


Photo

This was a tough one! - M101 from orange zone

  • Please log in to reply
16 replies to this topic

#1 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 17 May 2013 - 02:35 PM

I'm fairly happy with how much i was able to squeeze out of this, given I shot from my suburban back yard, and the object was clear of trees for only about an hour. Nevertheless I had to take advantage of better-than-usual transparency and the rare cool, dry night in Flow-ree-duh this time of year.

Comments and critiques welcome, of course, and thanks for looking.

Imaging scope: ST120
Camera: SS G3 OSC w/ Baader Semi-APO filter
Mount: CG5
Guide: SSAG/ST80/PHD
Subs: 13 X 300s
Capture: Nebulosity 3
Processing: Neb3 (stack), FITS Liberator (stretch), PS CS6

Attached Files



#2 Al Miller

Al Miller

    Apollo

  • *****
  • Posts: 1314
  • Joined: 25 Aug 2008
  • Loc: South, Texas

Posted 17 May 2013 - 03:09 PM

That's pretty impressive actually. The views on other monitors a likely different but, from what I see here you can pull more detail from this image. A bit of unsharp masking to tighten the stars and fine details would help as well. Still it's quite good from such a site. :goodjob:

#3 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 17 May 2013 - 03:36 PM

Thanks Al, I’ll try that. Still learning my way around PS so not familiar yet with what techniques (and when) to use.

BTW, love your sig, a man after my own heart. Wife says if I want another scope I have to offload one my 14 guitars. May just have to get a new wife (JK!)

#4 vorkus

vorkus

    Viking 1

  • -----
  • Posts: 561
  • Joined: 18 Aug 2010
  • Loc: Pgh, PA USA

Posted 17 May 2013 - 04:11 PM

Good job. I'm also in an orange zone and have fun trying to get galaxies like M101.

#5 JJK

JJK

    Surveyor 1

  • -----
  • Posts: 1866
  • Joined: 27 Apr 2008

Posted 17 May 2013 - 06:48 PM

Have you tried a LP filter? I recently got a Hutech LP2 for my SBIG STL-11000 one-shot color CCD camera, and it's made a difference in my Red Zone backyard.

The attached rough draft of M101 is a very tight crop of 5x600s light subs w/no flat field correction (- 20*C) I took last weekend (Ceravolo 300 @ f/4.9). The H-II regions are apparent. The color balance and color noise will get better when I use flats and take a lot more images. The stars will look tighter when I tweak the collimation and use the stacking routine correctly (for some reason, the program didn't work as I expected in this round of imaging).

Attached Files



#6 CounterWeight

CounterWeight

    Cosmos

  • *****
  • Posts: 8101
  • Joined: 05 Oct 2008
  • Loc: Palo alto, CA.

Posted 17 May 2013 - 11:28 PM

Serge,

Taking a quick look at the data as posted it seems pretty clipped in your initial stretching, the background shouldn't be that black? On the R/H side of the histogram you see it's sharp and vertical. If you are less agressive it will allow for more low contrast detail. When i push the levels there is a very sharp cutoff which to me means is too agressive near the black point?

Attached Files



#7 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 18 May 2013 - 08:03 AM

Hi Jim - I really appreciate your making the time and effort to help out. Here's a screen cap of my initial stretch of the linear FITS file I get from stacking in Neb3, after cropping stacking artifacts. I'm clipping the white just to show up the noise and LP gradient I'm dealing with; I'd normally leave the white point at maximum before bringing the stretched tif into PS.

You can see there's a distinct and well-separated black peak to the left of the main histogram peak. For the image in the OP I had set the black level between those 2 peaks, assuming the leftmost peak was due to some stacking artifact that I hadn't quite cropped. I'll try re-processing without clipping out that leftmost peak and see what happens.

I'd appreciate any other tips and suggestions, and thanks again for your help.

Attached Files



#8 JJK

JJK

    Surveyor 1

  • -----
  • Posts: 1866
  • Joined: 27 Apr 2008

Posted 18 May 2013 - 09:31 AM

I'd be curious to see the result of reprocessing the image. Light pollution and LP gradients complicate getting the color balance correct. I still think you'd benefit from a filter.

I made a typo in my op. The filter I purchased was a Hutech LPS2.

#9 DNTash

DNTash

    Viking 1

  • *****
  • Posts: 992
  • Joined: 01 Sep 2007
  • Loc: GMT+2

Posted 18 May 2013 - 09:49 AM

Very nice through the ST120.

#10 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 18 May 2013 - 11:11 AM

I'd be curious to see the result of reprocessing the image. Light pollution and LP gradients complicate getting the color balance correct. I still think you'd benefit from a filter.

I made a typo in my op. The filter I purchased was a Hutech LPS2.

Thanks for the info. I'll post my next re-process attempt later. I have a broadband LP filter I use for visual, but find it gives CCD images a blue-green cast that's hard to correct. I'll check out the Hutech. Another one I was looking at was the Orion SkyGlow Astrophotography filter.

#11 CounterWeight

CounterWeight

    Cosmos

  • *****
  • Posts: 8101
  • Joined: 05 Oct 2008
  • Loc: Palo alto, CA.

Posted 18 May 2013 - 11:35 AM

Serge, thanks for posting that FITS image. Very sharp cutoff there with that void between it and the remainder of that signal to the R/H side. So maybe that is more just an artifact of how the CCD is biased or signal balanced. M101 is really tough IMO because so much of it is very low luminance and it's signal so near the noise level. You can see what I'm referring to in the image John (JJK) posted.

I also image in heavy LP and it makes RGB a headache at best. I also try to use the IDAS-LP2 for it, but have problens with my current setup making that happen. There is also HLVG and Gradient Exterminator for helping in post processing. When I was doing OSC I always split the color channels into their own color panes and went after things sepirately, but you have to be very careful with desired signal and the balance when you reassemble. There are also things to try jsut with balance background signal or normalize it before appling initial stretching, but also IMO tricky. With my current mono setup I just take a lot of subs at smaller exposure length, but then I am in a white zone for LP.

Just sort of rambling here, hope any of it is helpful.

.

#12 JJK

JJK

    Surveyor 1

  • -----
  • Posts: 1866
  • Joined: 27 Apr 2008

Posted 18 May 2013 - 11:47 AM

I'd be curious to see the result of reprocessing the image. Light pollution and LP gradients complicate getting the color balance correct. I still think you'd benefit from a filter. It's always better to spend time processing the best data you can obtain.

I made a typo in my op. The filter I purchased was a Hutech LPS2.

Thanks for the info. I'll post my next re-process attempt later. I have a broadband LP filter I use for visual, but find it gives CCD images a blue-green cast that's hard to correct. I'll check out the Hutech. Another one I was looking at was the Orion SkyGlow Astrophotography filter.


I know what you mean. I currently use Nebulosity to correct for the background color offset (blue cast before correction). I'm going to give PixInsight a try soon (it might deal with color offsets better).

FWIW, the Hutech LPS2 filter worked well for me, and I'm 22 miles NW of the White House (the improvement will be even better when I use flat field correction). I'm going to eventually get a FLI 16803 monochrome camera and narrow band filters, but those won't be much help with imaging galaxies and stars.

#13 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 18 May 2013 - 03:11 PM

OK, here's a re-process attempt. I think I squeezed out some more detail, especially in the core, but somehow I threw off the color balance I had in the OP. Jury's still out (for me) as to which I like better. Would love to hear from you folks.

Thanks again for all who've commented and made suggestions. You're all a big help. Whoever said this was easy (if anybody ever did) should be beaten over the head with a fork and wedge.

Attached Files



#14 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 6034
  • Joined: 14 Jan 2013

Posted 18 May 2013 - 03:52 PM

Hi Serge:

There is definitely more detail than you are bringing out. Also the color balance is now too purple, although I prefer it to the original.

Here is a modified version. I can't get the color balance right with just the version posted. I'll take this down if you want me to.

If you post the unstretched version on dropbox or something, I wouldn't mind taking a quick try at it.

Attached Files



#15 SergeC

SergeC

    Viking 1

  • *****
  • Posts: 999
  • Joined: 23 Mar 2012
  • Loc: Gainesville, FL

Posted 18 May 2013 - 04:43 PM

...If you post the unstretched version on dropbox or something, I wouldn't mind taking a quick try at it.

Don't mind if you do. The following are Dropbox links:

My workflow is basically to stretch the RGB channels separately in FITS Liberator and then bring them together in PS. I'll do a number iterations of curves and levels, blurring, saturation (Adam Block method), etc. to bring up and denoise the color info. I use the synthetic Luminance for detail, so I apply deconvolution, curves and levels tweaking, gradient correction, etc. to that and then bring it on top of the united RGB.

If any of you do play around with my raw files, all I ask is that you jot down your workflow so I can try to reproduce it.

Thanks, and clear skies.

#16 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 6034
  • Joined: 14 Jan 2013

Posted 18 May 2013 - 06:02 PM

I'm having no success with the color data in this at all. The differences between channel are just very thin. I did dig more out of the luminosity channel, at the risk of being somewhat lumpy.

Processing was in PixInsight:

1) I brought it into PixInsight
2) Extracted the Luminosity Channel
3) Dynamic Background Extraction
4) Histogram Stretch
5) Wavelets
6) TGV Denoise
7) Histogram Stretch

Attached Files



#17 JJK

JJK

    Surveyor 1

  • -----
  • Posts: 1866
  • Joined: 27 Apr 2008

Posted 18 May 2013 - 09:51 PM

...If you post the unstretched version on dropbox or something, I wouldn't mind taking a quick try at it.

Don't mind if you do. The following are Dropbox links:

My workflow is basically to stretch the RGB channels separately in FITS Liberator and then bring them together in PS. I'll do a number iterations of curves and levels, blurring, saturation (Adam Block method), etc. to bring up and denoise the color info. I use the synthetic Luminance for detail, so I apply deconvolution, curves and levels tweaking, gradient correction, etc. to that and then bring it on top of the united RGB.

If any of you do play around with my raw files, all I ask is that you jot down your workflow so I can try to reproduce it.

Thanks, and clear skies.


Serge, what does the data look like if you don't split the RGB?

In my very rough draft of M101, I used Nebulosity 2.x to dark subtract, DeBayer to color, align & stack (I goofed up this part), cropped, remove color background offset, save to TIFF. I then used PhotoShop (curves & level, mild add-on Topaz Denoise filter, Neil Carboni's hard gradient removal, and weak Smart Sharpen). I didn't split the color channels.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics