Jump to content


Photo

M33 - LRGBHa Data for Processing - No Copyright

  • Please log in to reply
53 replies to this topic

#26 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 5484
  • Joined: 20 Jul 2007
  • Loc: California

Posted 17 November 2012 - 01:25 AM

Indeed, 2x2 binned RGB will not work as well for making Synthetic Luminance.
I only use 1x1 binning for RGB. I did try some 2x2 a few times, but I found that the data just was not as good as 1x1 for my purpose. I just am not a fan of 2x2 and then adding it to 1x1 luminace data.
Blueman

Hi Dave,

Yes, it was for testing purpose to compare "real" Luminance with "Synthetic" Luminance. To an average person, it was difficult to tell the difference but when I was comparing the two by blinking in between the two, "real" Luminance was the winner.

It's possible if the original RGB subs were capture in binned 1x1, the result would have been different but you would have to capture significantly more RGB subs to nearly match with "real" Luminance. That possibly explains why Synthetic Luminance was not as sharp due to RGB being captured in binned 2x2.

Peter



#27 rigel123

rigel123

    Voyager 1

  • *****
  • Posts: 11288
  • Joined: 29 Jun 2009
  • Loc: SW Ohio

Posted 17 November 2012 - 02:01 AM

It's too bad you don't have clear enough skies to capture the luminance data. I think it is most important.

I remember something about Adam Block showing that one could take RGB data and apply a heavy gaussian blur to it and when layered with the crisp 1x1 binned luminance data would produce a beautiful color image.


That is how I work on most of my images. Most of the time goes to luminance and less time on RGB which I blur pretty heavily and then add the Luminance to it. In narrowband I typically follow JP Metsavainio's Tone Mapping procedure where you remove the stars in the individual channels, stretch and blur the data, combine them for your chrominance, and then add the Luminance to that. It's great for keeping noise down in an image.

#28 SL63 AMG

SL63 AMG

    Viking 1

  • *****
  • Posts: 864
  • Joined: 21 Dec 2009
  • Loc: Williamson, Arizona

Posted 17 November 2012 - 03:28 AM

Ok Warren, you have been selected to come to Arizona and teach me.

Free Bed and Breakfast and use of the 18" DOB.

#29 StevenMx

StevenMx

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 11 Jun 2012
  • Loc: Connecticut, USA

Posted 17 November 2012 - 09:30 AM

Hi Dave,
I am with Peter Reno, I don’t usually use auto STF except to get a quick look at a linear image. When I use it I usually deselect the linked channels. I have dropped the STF process onto HST but rarely do I leave it at that setting before applying it. I do my best to manipulate the individual channels to keep the colors balanced “flat” before applying. Otherwise you end up with an unnatural (in relation to the RGBs) color in the end. If your histogram flattens “like a soft bump in the road” it might be because you forgot to disable STF before applying it to HST. You asked me about gradients, not sure I can answer that one. You have much more experience imaging than I do. And I only have experience in one shot color ccd.
Btw...my version of your M33 got image of the day today on AstroBin. That’s kind of exciting, I’ll take it!! :jump:

#30 pfile

pfile

    Gemini

  • -----
  • Posts: 3169
  • Joined: 14 Jun 2009

Posted 17 November 2012 - 12:18 PM

sometimes with the black bars left in the image, or because of a lot of background noise, the histogram appears empty. this is because there's a zillion pixels at 0 and the huge spike at 0 is hiding the rest of the data - the scale of the histogram displays has been adjusted to show the spike, but then you can't see the "real" data.

either you can crop, or do some background noise reduction, or just use the histogram zoom tools. there are 4 little boxes with "1"s in them between the bottom and top histogram displays. they control the horizontal and vertical zoom for the bottom and top displays. i've had to zoom in 500-600x sometimes because of the noise/black bar spike!

#31 Peter in Reno

Peter in Reno

    Fly Me to the Moon

  • *****
  • Posts: 6190
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 18 November 2012 - 07:59 PM

Dave,

In case you didn't see it, I posted my process of Bubble Nebula at your older thread at:

http://www.cloudynig...5475418/page...

I also explained why I was wrong when I said that there's nothing wrong with the blue image of Bubble Nebula.

Peter

#32 stephen63

stephen63

    Viking 1

  • *****
  • Posts: 547
  • Joined: 19 May 2010
  • Loc: Central Pa

Posted 20 November 2012 - 02:04 AM

Dave,
My latest attempt at your data. I use a laptop to process everything so perhaps that is the reason my images appear off. I have calibrated my monitor but who knows if its right. Typically, I download an image of the object from a reputable photographer, and bounce the color values against what I'm working on. In this case, I tried to match the colors of Robert Gendler's version of this image. Specifically, I tried to increase the detail of NGC 604.
As to your filter,
The left side of the galaxy that seems to be missing data is just that, a lack of star formation that is probably due to its interaction with another galaxy, perhaps, M31. Or, its a dust cloud. Either way, it has nothing to do with your filter position during capture.
I replaced your original with this.
http://www.flickr.co...in/photostream/

#33 StevenMx

StevenMx

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 11 Jun 2012
  • Loc: Connecticut, USA

Posted 20 November 2012 - 04:06 PM

Hi Dave, here is an updated version without Decon artifacts.

Daves' Larger M33

Attached Files



#34 StevenMx

StevenMx

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 11 Jun 2012
  • Loc: Connecticut, USA

Posted 24 November 2012 - 02:26 PM

Hi Dave, I started again from beginning using a couple different masks. This is my final version, hope you like it.
Daves' Larger M33

Attached Files



#35 manoss27

manoss27

    Lift Off

  • -----
  • Posts: 5
  • Joined: 25 May 2013
  • Loc: GREECE

Posted 24 October 2013 - 03:47 PM

thank you for the files!!!!

#36 SL63 AMG

SL63 AMG

    Viking 1

  • *****
  • Posts: 864
  • Joined: 21 Dec 2009
  • Loc: Williamson, Arizona

Posted 26 October 2013 - 07:26 PM

You are quite welcome.

I'll post more soon. I am gathering lots of data on different objects with an RCOS 12.5" and 16803 chip.

#37 coinboy1

coinboy1

    Messenger

  • *****
  • Posts: 467
  • Joined: 03 May 2011
  • Loc: Tulsa, OK

Posted 28 October 2013 - 05:48 PM

Yes thanks for the good data. I am trying my hand at some different processing software. This is very good practice for me.

#38 alpal

alpal

    Soyuz

  • -----
  • Posts: 3694
  • Joined: 15 Jun 2009
  • Loc: Melbourne Australia.

Posted 28 October 2013 - 08:34 PM

Post deleted by alpal

#39 alpal

alpal

    Soyuz

  • -----
  • Posts: 3694
  • Joined: 15 Jun 2009
  • Loc: Melbourne Australia.

Posted 28 October 2013 - 11:15 PM

Thanks Dave,
It's nice to get such wonderful data to work with.
I think Steve's version of larger M33 is the best so far.
This is my take on it after a stretch function x^1/5 in FITs Liberator & processing in PS.
I used ArcSin stretch for the core & blended with masks.
My aim was to try & bring out the faint areas without overexposing the brighter.
This was my 2nd go at it - I deleted the first one.

Larger version here:
http://www.flickr.co...703135/sizes/o/

Attached Files



#40 Joe F Gafford

Joe F Gafford

    Vanguard

  • -----
  • Posts: 2088
  • Joined: 15 Dec 2006
  • Loc: Denver, Colorado, US

Posted 28 October 2013 - 11:52 PM

Thanks for the tryout using another source than mine. Here's my submission. Maxim DL, CCD Soft, PS CS-6, Fits liberator.

Joe

Attached Files



#41 Ron (Lubbock)

Ron (Lubbock)

    Messenger

  • -----
  • Posts: 447
  • Joined: 17 Aug 2012
  • Loc: West TX

Posted 29 October 2013 - 12:39 PM

Here's an LRGB version without any H-alpha. I processed the image using only free software and shareware. I only spent about 20 minutes on it, so I'm sure that someone could easily do better.

IRIS was used for LRGB combination, Regim by Andreas Roerig was used for color calibration and gradient removal, and Deep Sky Stacker was used for stretching. Thanks to the OP for sharing these data, and many thanks to the generous people who make versions of their processing programs available to the public at no charge.

Posted Image

#42 SL63 AMG

SL63 AMG

    Viking 1

  • *****
  • Posts: 864
  • Joined: 21 Dec 2009
  • Loc: Williamson, Arizona

Posted 29 October 2013 - 07:01 PM

Alpal, the M33 image you created is very nice.

#43 alpal

alpal

    Soyuz

  • -----
  • Posts: 3694
  • Joined: 15 Jun 2009
  • Loc: Melbourne Australia.

Posted 29 October 2013 - 07:52 PM

Thanks Dave,
It's a great target & your data was excellent.
I'm just working on your NGC 1491 & NGC 281.

#44 manoss27

manoss27

    Lift Off

  • -----
  • Posts: 5
  • Joined: 25 May 2013
  • Loc: GREECE

Posted 04 December 2013 - 07:01 PM

My attempt without Ha
processing with pixinsight
http://www.flickr.co...is/11213443496/


#45 Dr Benway

Dr Benway

    Viking 1

  • *****
  • Posts: 657
  • Joined: 27 Mar 2008
  • Loc: Cherry Hill, Arkansas

Posted 05 December 2013 - 01:53 PM

Full resolution image

My take on your M33 data. Processed in PixInsight and Photoshop CS6. Looking forward to heading out west to see you in April!

John

#46 sc hokie

sc hokie

    Sputnik

  • -----
  • Posts: 33
  • Joined: 16 Aug 2009
  • Loc: Greenville, SC, USA

Posted 06 December 2013 - 09:05 AM

Threads like this really show you that there is no one "right" way of processing. Same data and so many varied outcomes!

#47 neptun2

neptun2

    Viking 1

  • -----
  • Posts: 992
  • Joined: 04 Mar 2007
  • Loc: Bulgaria

Posted 06 December 2013 - 09:50 AM

Here is my attempt also.

Posted Image

Very good data Dave. Thanks for sharing it. Btw i see that you used mean method for stacking the images. In the upper left corner there is something looking like a satellite trail in the blue channel. Is that correct? If so you can use sigma clip stacking method to completely remove such trails. It works very good if you have more than 6 or 7 frames. If they are less maximdl have also slightly modified version of that method called SD mask which work fine vene with 4 frames and removes satellite trails. Clear skies.

#48 StevenMx

StevenMx

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 11 Jun 2012
  • Loc: Connecticut, USA

Posted 06 December 2013 - 06:16 PM

I processed this same M33 data about a year ago. I thought I would revisit it and apply things I have learned since last year.

Ha_LRGB

Larger Image
Posted Image

Attached Files



#49 StevenMx

StevenMx

    Explorer 1

  • -----
  • Posts: 85
  • Joined: 11 Jun 2012
  • Loc: Connecticut, USA

Posted 06 December 2013 - 06:41 PM

Threads like this really show you that there is no one "right" way of processing. Same data and so many varied outcomes!


I think there is definitely a "right" and wrong way to process astronomy images. But there is certainly a creative license that comes with it in the end.

#50 Bradicus

Bradicus

    Mariner 2

  • -----
  • Posts: 293
  • Joined: 30 Nov 2013
  • Loc: Southwest Idaho

Posted 07 December 2013 - 01:08 AM

Since I've been thinking about jumping into CCD imaging, I thought I would give this a try. Turned out to be easier than I thought using my favorite image acquisition software -- Nebulosity. This only took 4 steps:

1. Batch align the 4 individual channels (LRGB)
2. LRGB Color Synthesis using the Traditional HSI option
3. Auto color balance
4. Digital Development (DDP) using default settings

Each of these steps are totally automated.

I then just cropped the image and saved it as a JPEG. There is a lot more that can be done in Photoshop, but I thought it was pretty cool how much could be done with this data in just a couple clicks.

Attached Files








Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics