Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Rosette Nebula - Someone want to have a go at this data

  • Please log in to reply
41 replies to this topic

#1 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 16 January 2022 - 10:45 PM

This is about 9 hours worth of data taken over 4 nights with two different telescopes and two different cameras.  There's 50 x 180 sec subs at ISO800 taken thru an AT80ED using a .8x reducer with a Ha modded Canon T4i.  The second night was with the same combo except different camera.  I took 55 x 180 sec subs at 150 gain with my new to me Orion Starshoot G16 (same as ASI-1600MC) using the same scope and reducer and no filters other than a UV/IR cut filter.  Third night was with the same OSC using an Orion ED80 and Orion flattener.  Nearly the same amount of subs (60 x 180) but I bumped up the gain to 200 and used an Orion Imaging Skyglow filter.  The fourth night was taken with the same ED80 and .8x reducer with an Optolong UHC filter as the moon was very bright.  

 

I did sort of a mosaic stack with APP and trimmed out about 45 mins of subs due to clouds and some not so round stars after my telescope passed the meridian and the tracking got a little jerky due to poor balance.  I'd like for someone who's good at processing to see what they can do with this integration.  My processing skills are not great, especially with nebula and I think there's some decent data here.  Here's a link to the unedited stacked image that has had nothing done except for a crop.  

 

This was the result of just doing a light pollution correction and crop in APP.  I did no denoising.  The colors with APP seemed a little redder, for lack of a better description.  I took the same stack and did a channel extraction, linear fit, and RGB combination in PI to balance the colors and then a DBE to get rid of some of the gradients.  Still did no denoising but colors looked tamer.  

 

Have a go at it if you like, and let me know how you ended up with your results.  Thanks!

 

Attached Thumbnails

  • Rosette_APP_LPC.JPG
  • Rosette_PI_DBE_SCNR.JPG

  • Dren, 72Nova, Tim J Fowler and 2 others like this

#2 rj144

rj144

    Apollo

  • -----
  • Posts: 1,297
  • Joined: 31 Oct 2020

Posted 16 January 2022 - 11:27 PM

Quick 10 minute edit with Startools and Affinity using Ritson macros:

 

Rosette-composite-session-1-session-2-se

 

 

I'm jealous....what Bortle zone was this?  I shot 9.5 hours of the Rosette and my data was no where near this clean in Bortle 8.  Nice data!

 

Here's my data before post processing:

 

rose-bw-4tiff.jpg


Edited by rj144, 16 January 2022 - 11:48 PM.

  • jasonjeremiah, JerryStellar and Tim J Fowler like this

#3 IsaacT1990

IsaacT1990

    Lift Off

  • -----
  • Posts: 11
  • Joined: 27 Sep 2021

Posted 16 January 2022 - 11:34 PM

This is a small resizing of a quick edit through mostly photoshop. I did levels stretching and used levels to balance the histogram. Then used StarXTerminator to separate into Starless and Starry. Then took starless layer over to PixInsight for a quick Automatic Background Extraction. Then pulled that back over into the photoshop document. Used levels and curves to stretch it a few more times. Used HLVG to get rid of a little green hue in the shadows. Added the stars back in. Took about 10 minutes tops. No noise reduction was used. Once the background extraction was done, the nebulosity was very eager to come to the forefront.

Attached Thumbnails

  • RosetteChallenge-smaller.jpg

  • Steve_M_M, jasonjeremiah, 72Nova and 2 others like this

#4 Sky King

Sky King

    Viking 1

  • *****
  • Posts: 988
  • Joined: 16 Mar 2017
  • Loc: Arizona

Posted 17 January 2022 - 12:14 AM

Nice image! Here's a few minutes in Affinity Photo with Prodigital Software's AstroFlat Pro filter. 

 

redo2.png


Edited by Sky King, 17 January 2022 - 04:19 PM.

  • jasonjeremiah and JerryStellar like this

#5 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 28,344
  • Joined: 27 Oct 2014

Posted 17 January 2022 - 12:35 AM

Quick hack job in PI.  Better version than the crummy (required) CN jpg here.

 

https://www.astrobin.com/cyk5qu/

 

Suggestion.  Even modestly bright stars have large white cores.  You might try half as much subexposure (and twice the number of subs, of course).

 

anothers Rosette.jpg


Edited by bobzeq25, 17 January 2022 - 12:37 AM.

  • jasonjeremiah, JerryStellar, 72Nova and 2 others like this

#6 Dren

Dren

    Mariner 2

  • *****
  • Posts: 292
  • Joined: 09 Jan 2020
  • Loc: San Diego, CA

Posted 17 January 2022 - 01:05 AM

Here's my try at your data.  A bit darker than others.  You can see dark circles around the stars, which are a remnant of Starnet.  Didn't spend the time to touch those up.  One of the things I really enjoy about this hobby is how we can end up with so many different results from the same data.  Thanks for sharing.  

 

Rosette1600.jpg


  • Traveler, DesertCrawler, CharLakeAstro and 2 others like this

#7 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 01:22 AM

I'm jealous....what Bortle zone was this?  I shot 9.5 hours of the Rosette and my data was no where near this clean in Bortle 8.  Nice data!

 

Thanks!  Actually, my skies are not much better than yours.  I live on the outskirts of Charleston SC and have a street lamp at the end of the driveway I image from.  Around Brtle 6 - 7.  



#8 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 01:24 AM

This is a small resizing of a quick edit through mostly photoshop. I did levels stretching and used levels to balance the histogram. Then used StarXTerminator to separate into Starless and Starry. Then took starless layer over to PixInsight for a quick Automatic Background Extraction. Then pulled that back over into the photoshop document. Used levels and curves to stretch it a few more times. Used HLVG to get rid of a little green hue in the shadows. Added the stars back in. Took about 10 minutes tops. No noise reduction was used. Once the background extraction was done, the nebulosity was very eager to come to the forefront.

That looks great.  Actually looks like a rose.  Good job!



#9 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 01:29 AM

Quick hack job in PI.  Better version than the crummy (required) CN jpg here.

 

https://www.astrobin.com/cyk5qu/

 

Suggestion.  Even modestly bright stars have large white cores.  You might try half as much subexposure (and twice the number of subs, of course).

 

attachicon.gifanothers Rosette.jpg

I'm still figuring out this new camera and what settings to use.  I did the Sharpcap analysis, but the smart histogram refuses to work with the camera, so gain and offset has been a crap shoot.  I believe I will dial back the gain to 150 (100 is zero gain in this camera) and cut the exposure times in half and see what that gives me.  Full well is only 20k but I'm really enjoying shooting with it.  Definitely much less noise than my Canon T4i.  I have an L-enhance on the way.  My try to add some narrowband to this integration and work on my processing.



#10 Mike in Rancho

Mike in Rancho

    Vanguard

  • -----
  • Posts: 2,145
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 17 January 2022 - 02:40 AM

Hi Jeremiah, thanks for posting your link.

 

Took a try at it.  I'm not sure a combo stack of all those different cameras, telescopes, and filters is going to do you any favors.  Compared to sticking to one setup and racking up your 9 hours that way.

 

But your result seems reasonable and it's better than any Rosette I've done. smile.gif

 

With like 4 different color profiles in here, I didn't even bother trying to set the balance via star sampling, just kind of played with it by eye until I liked the contrast.  There's some decent details you have here in those darker shadowy shapes, and of course the big "Y."

 

gallery_345094_15786_399723.jpg


  • jasonjeremiah, CharLakeAstro, JerryStellar and 3 others like this

#11 F.Meiresonne

F.Meiresonne

    Cosmos

  • *****
  • Posts: 8,516
  • Joined: 22 Dec 2003
  • Loc: Eeklo,Belgium

Posted 17 January 2022 - 03:17 AM

My take in Startools and a bit of tweaking in GIMP.

 

Later i will try a HOO approach. The orange is not really my thing.

 

Nice data though

Attached Thumbnails

  • RosettaJason (Large).jpg

Edited by F.Meiresonne, 17 January 2022 - 03:36 AM.

  • jasonjeremiah, JerryStellar, 72Nova and 2 others like this

#12 acrh2

acrh2

    Apollo

  • -----
  • Posts: 1,033
  • Joined: 16 Mar 2021

Posted 17 January 2022 - 03:38 AM

51825861598_67ea5ca32a_o.jpg


  • F.Meiresonne, Steve_M_M, Traveler and 8 others like this

#13 Tim J Fowler

Tim J Fowler

    Viking 1

  • *****
  • Posts: 932
  • Joined: 21 Jul 2020
  • Loc: Seneca, South Carolina

Posted 17 January 2022 - 09:36 AM

Nice data, thanks for sharing! I gave a quick run through Siril (photometric color calibration, remove green noise, deconvolution, asinh transformation, histogram transformation and then finally background extraction), and it looked pretty good straight out of Siril, but I wanted to make it a little more vibrant so I took it into GIMP for a little levels and curves adjustment. I still thought the color was a bit off, though really who is to say what the true colors really are. I'm just not a huge fan of images produced by Ha modified cameras, and I'm still not sold on the colors I achieved here, and the stars are a little soft, but this was literally about 5 minutes worth of processing. I think I would have to run it through Starnet++ to achieve better results with the stars and overall color.

Attached Thumbnails

  • Siril 2 GIMP Rosette_composite-session_1_session_2_session_3_session_4-mod Siril GIMP.jpg

  • jasonjeremiah and JerryStellar like this

#14 72Nova

72Nova

    Supernova

  • -----
  • Posts: 384
  • Joined: 22 Feb 2020
  • Loc: Coachella Valley, CA

Posted 17 January 2022 - 09:54 AM

Here is my attempt at your data using Astro Pixel Processor.  For what it is worth, I ran a few more iterations in the light pollution tool, then calibrated star colors (this changed the color of the nebula), and then ran the star reduction tool (reduced stars to 50%).

 

Image was stretched in APP at 15% and then I cropped, adjusted contrast/brightness and boosted saturation in Affinity. 

 

I have an AT80ED that I want to start using for AP and it looks like it is capable of producing good images.

 

Thanks for sharing!  

Attached Thumbnails

  • Rosette_composite_resize.jpg

Edited by 72Nova, 17 January 2022 - 10:38 AM.

  • F.Meiresonne, jasonjeremiah, bobzeq25 and 1 other like this

#15 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 10:45 AM

My take in Startools and a bit of tweaking in GIMP.

 

Later i will try a HOO approach. The orange is not really my thing.

 

Nice data though

Thanks, Freddy.  If you'd like I could share the individual mosaic panels if that would help with the HOO approach.

 

Panel with no filter taken with Ha modded T4i

Panel with no filter taken with Orion Starshoot G16

Panel with Orion Skyglow Imaging filter and Starshoot G16 

Panel with Optolong UHC and Starshoot G16

 

Thanks for taking the time to process my data - 

 

Jeremiah


  • Mike in Rancho likes this

#16 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 10:51 AM

Hi Jeremiah, thanks for posting your link.

 

Took a try at it.  I'm not sure a combo stack of all those different cameras, telescopes, and filters is going to do you any favors.  Compared to sticking to one setup and racking up your 9 hours that way.

 

But your result seems reasonable and it's better than any Rosette I've done. smile.gif

 

With like 4 different color profiles in here, I didn't even bother trying to set the balance via star sampling, just kind of played with it by eye until I liked the contrast.  There's some decent details you have here in those darker shadowy shapes, and of course the big "Y."

 

gallery_345094_15786_399723.jpg

Thanks, Mike.  The different telescope/camera combinations wasn't necessarily by design.  A couple of nights were last minute setups as it was supposed to be cloudy and turned out clear, and I just put what telescope I happened to have the Losmandy D plate on (just recently moved to a CGEM DX) and it isn't the easiest to switch around last minute.  It's an ironwood dovetail I carved myself.  I've also been testing a new OSC camera I bought recently and playing with different settings.

 

Having said all of that, I really know too little about astrophotography and the conventional do's and don't s to adhere to them.  Every once in a while, the results turn out fairly decent!


  • Mike in Rancho likes this

#17 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 10:56 AM

51825861598_67ea5ca32a_o.jpg

Wow!  I love what you did here.  Would you mind sharing your processing flow?  I'll try to follow along.  Thanks!

 

Jeremiah



#18 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 10:59 AM

Nice data, thanks for sharing! I gave a quick run through Siril (photometric color calibration, remove green noise, deconvolution, asinh transformation, histogram transformation and then finally background extraction), and it looked pretty good straight out of Siril, but I wanted to make it a little more vibrant so I took it into GIMP for a little levels and curves adjustment. I still thought the color was a bit off, though really who is to say what the true colors really are. I'm just not a huge fan of images produced by Ha modified cameras, and I'm still not sold on the colors I achieved here, and the stars are a little soft, but this was literally about 5 minutes worth of processing. I think I would have to run it through Starnet++ to achieve better results with the stars and overall color.

Thanks, Tim.  Using the different filters probably didn't help with the colors, but if I don't use filters of some sort, it's nearly impossible to get decent signal over the light pollution here.  I'm surrounded by neighbors that refuse to turn off their porch lights and still have Christmas lights up and have a nasty LED street lamp at the end of my driveway.


  • Tim J Fowler likes this

#19 crackout

crackout

    Ranger 4

  • -----
  • Posts: 352
  • Joined: 12 Mar 2021
  • Loc: Augsburg, Germany

Posted 17 January 2022 - 11:34 AM

Here's my take.
Recently, I found myself using more and more pastelish subdued colour palettes. This edit makes no expection. smile.gif

 

Rosette Jeremiah

 

Like others have already pointed out: Some stars are blown out. You should dial down your subexposure time a little to fix this.

I also suggest to integrate the 'pure' RGB data separately in addition to add their stars in later for the best colour.

 

Link to the uncompressed TIFF: https://www.dropbox....remiah.tif?dl=0


Edited by crackout, 17 January 2022 - 11:44 AM.

  • F.Meiresonne, jasonjeremiah, CharLakeAstro and 3 others like this

#20 jasonjeremiah

jasonjeremiah

    Viking 1

  • *****
  • topic starter
  • Posts: 533
  • Joined: 28 Nov 2009
  • Loc: Hanahan SC

Posted 17 January 2022 - 11:43 AM

Here's my take.
Recently, I discovered, I like the more pastelish subdued style a lot. This edit makes no expection. smile.gif

 

 

 

Like others have already pointed out: Some stars are blown out. You should dial down your subexposure time a little to fix this.

I also suggest to integrate the 'pure' RGB data separately and add those stars in later for the best colour.

 

Link to the uncompressed TIFF: https://www.dropbox....remiah.tif?dl=0

Wow, this may be my new favorite.  I'll drop the exposure times and gain.  Going to try for the Jellyfish tonight so I'll try 150 gain and 90 second exposures.  That is if the howling winds around here will let up.



#21 crackout

crackout

    Ranger 4

  • -----
  • Posts: 352
  • Joined: 12 Mar 2021
  • Loc: Augsburg, Germany

Posted 17 January 2022 - 11:46 AM

Wow, this may be my new favorite.  I'll drop the exposure times and gain.  Going to try for the Jellyfish tonight so I'll try 150 gain and 90 second exposures.  That is if the howling winds around here will let up.


Glad you like it.

Do some test exposures and have a look at them without any preview stretch applied. If you can see the stars nice and bright, your exposure time is too long.
  • jasonjeremiah likes this

#22 acrh2

acrh2

    Apollo

  • -----
  • Posts: 1,033
  • Joined: 16 Mar 2021

Posted 17 January 2022 - 11:56 AM

Wow!  I love what you did here.  Would you mind sharing your processing flow?  I'll try to follow along.  Thanks!

 

Jeremiah

Siril Photometric Color Calibration, SCNR.

I cropped the data a bit because the edges looked to have some stacking artifacts in "Histogram" viewing mode.

Background extraction (took a while to get the sample boxes in the right places to have a decent gradient removal, because that gradient was kind of nasty.)

Then a Histogram Transformation with Autostretch - I usually adjust the left slider to get to about 0.100 of Shadows Clip(%), and the middle slider a bit to the right because autostretch often overdoes it a bit.)

Then saved it as a 16-bit tiff.

 

Then Starnet++ with default settings to remove stars.

 

Then Startools.

Open starry tiff, click Layer, Open starless tiff, Subtract Layer mode, Keep, Save as stars tiff.

Open starless tiff,

Contrast if it works (sometimes noise makes Contrast useless), HDR if it works (sometimes it makes things weird looking) - I didn't use HDR on your data.

Sharp (keep mask as is,) Deconvolution (works great without stars at Primary Radius of about 2-2.5).

Save as a tiff.

Then Topaz Denoise - there you have to play with settings for each individual image, but usually Standard or Severe Noise work great with about 80% of Recover original detail.

If you don't have Topaz, then Startools can do Noise Reduction as well when you deactivate Track.

 

Back to Startools.

Open denoised tiff, Layer, Open stars tiff, Layer Mode: Add, Brightness Mask Mode: where fg is dark use bg, Brightness Mask Power:  0.5-0.7, Keep, Save.

Then in Gimp, maybe a little Sharp mask, adjust some contrast, maybe curves with individual colors, saturation to taste. If the background has too much color you can clip that particular color in Levels. At that point, it's whatever looks good.


  • Traveler likes this

#23 acrh2

acrh2

    Apollo

  • -----
  • Posts: 1,033
  • Joined: 16 Mar 2021

Posted 17 January 2022 - 11:57 AM

Here's my take.
Recently, I found myself using more and more pastelish subdued colour palettes. This edit makes no expection. smile.gif

 

 

 

Like others have already pointed out: Some stars are blown out. You should dial down your subexposure time a little to fix this.

I also suggest to integrate the 'pure' RGB data separately in addition to add their stars in later for the best colour.

 

Link to the uncompressed TIFF: https://www.dropbox....remiah.tif?dl=0

What's your secret to making stars like that?


  • crackout likes this

#24 crackout

crackout

    Ranger 4

  • -----
  • Posts: 352
  • Joined: 12 Mar 2021
  • Loc: Augsburg, Germany

Posted 17 January 2022 - 12:05 PM

What's your secret to making stars like that?

I extract the stars after a minimal stretch to the image.

Then revert to a starful image and apply a thorough stretch, remove stars again, process further and add the minimal stars back in at the end. 
 

The minimal stretch is a combination of slight ArcsinhStretch and HistogramTransformation steps to retain colour and avoid bloat. This operation can be tricky and may require several attempts to get right. 


Edited by crackout, 17 January 2022 - 12:07 PM.

  • CharLakeAstro likes this

#25 Tim J Fowler

Tim J Fowler

    Viking 1

  • *****
  • Posts: 932
  • Joined: 21 Jul 2020
  • Loc: Seneca, South Carolina

Posted 17 January 2022 - 12:15 PM

And for my "can't leave well enough alone over the top" approach...

Attached Thumbnails

  • bold small BE Rosette_composite-session_1_session_2_session_3_session_4-mod.jpg

  • 72Nova likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics