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

Is M81 M82 really this hard or is it me?

Astrophotography Software
  • Please log in to reply
69 replies to this topic

#26 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 186
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 07 May 2021 - 10:23 PM

Here's my process of OP's data, processed as L+Synthetic L from RGB+RGB in startools. I think it looks pretty good, the corners indicate there's a spacing or flatness issue, and the HA data looked out of focus so I ignored it. There's also a slight shadow on the very right side, which is a little weird, but it didn't pose much problem and I mostly ignored it. I did notice a blue halo around pretty much every star, I'd guess focus was off a little, or maybe seeing degraded some vs the other channels. I corrected that using the Filter module.

 

NewComposite_2.jpeg

 

So..uhh..I guess it's just you. laugh.gif


Edited by copper280z, 07 May 2021 - 10:27 PM.

  • dswtan, Jim Waters, Sky King and 5 others like this

#27 BQ Octantis

BQ Octantis

    Aurora

  • *****
  • Posts: 4,757
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 07 May 2021 - 10:27 PM

A basic 10-minute ($0.21) stretch of the RGB with Levels in Photoshop CS5 ($600):

 

(Click for full size @ 25% scale.)

gallery_273658_12412_112554.jpg

 

The background noise disappears at 16.7% scale. So I'd revise my 8 hour total reco to 6 hours total. Per channel! You also need a field flattener to make the stars in the corners round.

 

Let me know if you need a stretch tutorial…it should work in Gimp.

 

BQ


Edited by BQ Octantis, 07 May 2021 - 10:31 PM.

  • brlasy1, DJL and psugrue like this

#28 DJL

DJL

    Viking 1

  • -----
  • Posts: 636
  • Joined: 15 Aug 2020

Posted 07 May 2021 - 10:40 PM

https://www.cloudyni...s-what-is-this/ - 30 x 5 minutes OSC + 14 x 5 minutes Ha, Bortle 6. WO GT81 + field flattener.



#29 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 07 May 2021 - 10:46 PM

Here's my process of OP's data, processed as L+Synthetic L from RGB+RGB in startools. I think it looks pretty good, the corners indicate there's a spacing or flatness issue, and the HA data looked out of focus so I ignored it. There's also a slight shadow on the very right side, which is a little weird, but it didn't pose much problem and I mostly ignored it. I did notice a blue halo around pretty much every star, I'd guess focus was off a little, or maybe seeing degraded some vs the other channels. I corrected that using the Filter module.

 

attachicon.gifNewComposite_2.jpeg

 

So..uhh..I guess it's just you. laugh.gif

Thank you so much! I am close to tears right now seeing your image. I did not use a flattener but I can add that back in. Obviously, I need to work on my processing but knowing that my hardware and capture is at least in the ballpark is a relief.

 

Kind Regards,

 

Patrick  


Edited by psugrue, 07 May 2021 - 10:59 PM.

  • brlasy1, DJL and Islander13 like this

#30 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 186
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 07 May 2021 - 10:57 PM

Thank you so much! I am close to tears now seeing your image. I did not use a flattener but I can add that back in. Obviously, I need to work on my processing but knowing that my hardware and capture is at least in the ballpark is a relief.

 

Kind Regards,

 

Patrick  

No problem, processing is tricky and takes at least as much practice as the capture portion. The raw data you captured is overall very good. I'd suggest playing with the demos of StarTools and APP, and watching a few tutorial videos.


  • psugrue likes this

#31 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 07 May 2021 - 11:26 PM

No problem, processing is tricky and takes at least as much practice as the capture portion. The raw data you captured is overall very good. I'd suggest playing with the demos of StarTools and APP, and watching a few tutorial videos.

Will do. Now you have proven the potential... I am happy to spend several months or more to figure this out. I worked at Microsoft in Redmond for a number of years so I am not too scared of a PC.

 

Respect,

 

Patrick



#32 Idkwhatsgoingon

Idkwhatsgoingon

    Sputnik

  • -----
  • Posts: 40
  • Joined: 10 Apr 2021

Posted 07 May 2021 - 11:28 PM

I got this in pixinsight, I'm still new to PI so its not perfect. I didn't use your Ha data

 

https://drive.google...iew?usp=sharing


Edited by Idkwhatsgoingon, 07 May 2021 - 11:37 PM.

  • psugrue likes this

#33 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 880
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 07 May 2021 - 11:54 PM

Here's mine, also just fiddling around in Startools.  Thanks for letting me play with it! waytogo.gif  There are in fact galaxies here, not just...you know, blob things.  M81 has spiral arms and M82 some red wisps.

 

I also ended up killing off blue fringes, though some of that might have been because I balanced it fairly blue heavy.

 

Everything has probably already been mentioned that I can think of - flattener for the corners, check focus.  Also I think with mono, the idea is to get more L than the color channels, which will help with detail.  This was the opposite here and by a good bit.  But a mono imager can correct me on that if I'm off base.

 

I actually tried to warp the corner stars back into circles, which worked well on the little ones, but not the bigger ones for some reason.  So I just canceled out of that and left it as-is.

 

gallery_345094_15786_266826.jpg


  • brlasy1 and psugrue like this

#34 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 08 May 2021 - 12:09 AM

Here's mine, also just fiddling around in Startools.  Thanks for letting me play with it! waytogo.gif  There are in fact galaxies here, not just...you know, blob things.  M81 has spiral arms and M82 some red wisps.

 

I also ended up killing off blue fringes, though some of that might have been because I balanced it fairly blue heavy.

 

Everything has probably already been mentioned that I can think of - flattener for the corners, check focus.  Also I think with mono, the idea is to get more L than the color channels, which will help with detail.  This was the opposite here and by a good bit.  But a mono imager can correct me on that if I'm off base.

 

I actually tried to warp the corner stars back into circles, which worked well on the little ones, but not the bigger ones for some reason.  So I just canceled out of that and left it as-is.

 

gallery_345094_15786_266826.jpg

 

Here's mine, also just fiddling around in Startools.  Thanks for letting me play with it! waytogo.gif  There are in fact galaxies here, not just...you know, blob things.  M81 has spiral arms and M82 some red wisps.

 

I also ended up killing off blue fringes, though some of that might have been because I balanced it fairly blue heavy.

 

Everything has probably already been mentioned that I can think of - flattener for the corners, check focus.  Also I think with mono, the idea is to get more L than the color channels, which will help with detail.  This was the opposite here and by a good bit.  But a mono imager can correct me on that if I'm off base.

 

I actually tried to warp the corner stars back into circles, which worked well on the little ones, but not the bigger ones for some reason.  So I just canceled out of that and left it as-is.

 

gallery_345094_15786_266826.jpg

Wow! Just wow. How do you guys do this?


  • brlasy1 and Mike in Rancho like this

#35 BQ Octantis

BQ Octantis

    Aurora

  • *****
  • Posts: 4,757
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 08 May 2021 - 01:58 AM

Turns out, the green and the blue channels are out of focus, with the blue worse than the green. This resulted in blurry galaxies and stars with red centers. Be sure to refocus periodically. I focus in between aperture changes—about once per hour.

 

You also captured a satellite in one of the L subs that didn't average out in the final. Use Kappa Sigma Clipping in DSS to get rid of those.

 

Here's a crack at improving the sharpness with LR-RGB:

 

(Click for full size @ 25% scale.)

gallery_273658_12412_88437.jpg

 

BQ


Edited by BQ Octantis, 08 May 2021 - 02:57 AM.

  • brlasy1, Brandon0616, psugrue and 1 other like this

#36 JamesTX

JamesTX

    Messenger

  • *****
  • Posts: 480
  • Joined: 27 Aug 2017
  • Loc: Texas

Posted 08 May 2021 - 02:18 AM

Here's my take on your data.

 

Some feedback.  As others mentioned.. focus is off slightly.. except for Ha which is off by a decent amount.  Definitely want to refocus for Ha.  I also recommend a lot more lum data.  I know there's a suggestion out there recommending more color than lum.. but in my experience you can get by with very little rgb data compared to lum.  You can experiment and see which way works best for you.

 

 

Attached Thumbnails

  • Image31_1600.jpg

  • brlasy1 and psugrue like this

#37 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 880
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 09 May 2021 - 12:51 AM

Wow! Just wow. How do you guys do this?

Well, what steps did you take with your processing?

 

The underlying concepts of what is being done are pretty much the same I think, regardless of platform.  But they may be done in different ways, with different names, perhaps with different order required, as well as different levels of being "in the weeds" - meaning how much manual vs. automated work is required.  Gimp will be fairly similar to PS, for which there are plenty of guides and tutorials out there.

 

I could post a detailed log of what I did, which could be replicated if you have Startools (or tweaked to your own liking of course), but from a more general overview standpoint, this is what was done with your files:

 

L, R, G, and B opened and combined, with a synthetic L frame created and added based on the weighting of your integration.

 

Superpixel binning (scaling) down to 25% of the original size, which improves SNR.

 

Cropped for stacking artifacts and/or desired FOV.

 

Gradient and vignetting clean-up.

 

Stretch.

 

Contrast enhancement, in which I also compressed the dynamic range of the background (shadows).

 

High dynamic range (HDR) enhancement.

 

Sharpening.

 

Deconvolution (recovers for atmospheric distortion).

 

Color balancing and saturation.

 

Star shrinking and halo fringe taming.

 

Combination background suppression, additional star shrinking, and preliminary denoise.

 

Final denoise.

 

Go back and tame a few more star halo fringes.

 

Save!

 

 

Whatever program you use, once you get used to it, all of that stuff will start becoming second nature to you and you will develop a standard routine.  Which may vary depending on the particular target.

 

There aren't as many tutorials, or available plugin tools written, for Gimp, but there's a few out there.  And while I like Gimp and still use it for various things, I didn't really enjoy it for astro processing and so got an astro-centric program designed for this sort of stuff.


  • psugrue likes this

#38 Tom62e

Tom62e

    Ranger 4

  • *****
  • Posts: 323
  • Joined: 25 May 2014
  • Loc: Scottsdale, AZ

Posted 09 May 2021 - 01:11 AM

What's your the light pollution like?  Download a trial version of PixInsight and give that a try.  Here's my image of M81/82 taken from a Bortle 8+ site - backyard.  100mm f/5.5 scope.  90 sec x 240 each subs processed with PixInsight.

 

 

50960606071_fea6578bc8_b.jpgM81 82 Area by Jim Waters, on Flickr

Jim - this is amazing.  Look at that starburst!
 



#39 DJL

DJL

    Viking 1

  • -----
  • Posts: 636
  • Joined: 15 Aug 2020

Posted 09 May 2021 - 01:20 AM

You can see Holmberg IX in some of the reprocessed versions. I learned about this when I posted my M81 / M82 and received comments on it. 


  • Mike in Rancho likes this

#40 BQ Octantis

BQ Octantis

    Aurora

  • *****
  • Posts: 4,757
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 09 May 2021 - 04:01 AM

I would put your M82 quality scale limit at 40%:

 

(Click for 100% scale.)

gallery_273658_12412_31275.jpg

 

BQ


  • psugrue likes this

#41 BQ Octantis

BQ Octantis

    Aurora

  • *****
  • Posts: 4,757
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 09 May 2021 - 05:04 AM

Here's the Ha added to it. 40% scale.

 

(Click for 100% scale.)

gallery_273658_12412_93955.jpg

 

BQ


  • Mike in Rancho and psugrue like this

#42 sbharrat

sbharrat

    Viking 1

  • -----
  • Posts: 595
  • Joined: 28 Nov 2020
  • Loc: NJ, USA

Posted 09 May 2021 - 06:47 AM

...

 

Superpixel binning (scaling) down to 25% of the original size, which improves SNR.

 

Cropped for stacking artifacts and/or desired FOV.

 

Gradient and vignetting clean-up.

 

Stretch.

 

...

 

Deconvolution (recovers for atmospheric distortion).

 

Color balancing and saturation.

 

...

 

 

Have I been doing this wrong? deconv comes after stretch? 



#43 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 186
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 09 May 2021 - 09:32 AM

Have I been doing this wrong? deconv comes after stretch? 

In StarTools deconv comes after stretch, in other tools it's deconv firrst.


  • Mike in Rancho and sbharrat like this

#44 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 880
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 09 May 2021 - 04:19 PM

You can see Holmberg IX in some of the reprocessed versions. I learned about this when I posted my M81 / M82 and received comments on it. 

Wow, I just learned something new also!  waytogo.gif

 

It's fairly faint in my reprocess and I didn't do anything to help bring it out.  In fact, I nearly healed it out of the image, on the assumption that it was most likely an overcorrected dust mote.  lol.gif

 

I checked Stellarium quickly and saw that no, it's actually real, so I refrained from rearranging the universe.  But I also just assumed it was a stray piece of the spiral, not a separate dwarf with a cool name like Holmberg IX!


  • DJL likes this

#45 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 09 May 2021 - 05:09 PM

In terms of light pollution, I am less than 5km away from the center of downtown Boise on a Bortal 6/7 border. 



#46 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 10 May 2021 - 05:50 PM

Well, what steps did you take with your processing?

 

The underlying concepts of what is being done are pretty much the same I think, regardless of platform.  But they may be done in different ways, with different names, perhaps with different order required, as well as different levels of being "in the weeds" - meaning how much manual vs. automated work is required.  Gimp will be fairly similar to PS, for which there are plenty of guides and tutorials out there.

 

I could post a detailed log of what I did, which could be replicated if you have Startools (or tweaked to your own liking of course), but from a more general overview standpoint, this is what was done with your files:

 

L, R, G, and B opened and combined, with a synthetic L frame created and added based on the weighting of your integration.

 

Superpixel binning (scaling) down to 25% of the original size, which improves SNR.

 

Cropped for stacking artifacts and/or desired FOV.

 

Gradient and vignetting clean-up.

 

Stretch.

 

Contrast enhancement, in which I also compressed the dynamic range of the background (shadows).

 

High dynamic range (HDR) enhancement.

 

Sharpening.

 

Deconvolution (recovers for atmospheric distortion).

 

Color balancing and saturation.

 

Star shrinking and halo fringe taming.

 

Combination background suppression, additional star shrinking, and preliminary denoise.

 

Final denoise.

 

Go back and tame a few more star halo fringes.

 

Save!

 

 

Whatever program you use, once you get used to it, all of that stuff will start becoming second nature to you and you will develop a standard routine.  Which may vary depending on the particular target.

 

There aren't as many tutorials, or available plugin tools written, for Gimp, but there's a few out there.  And while I like Gimp and still use it for various things, I didn't really enjoy it for astro processing and so got an astro-centric program designed for this sort of stuff.

I have StarTools installed now but there are so many parameters. Please post your Startools log if you would be so kind.

 

Regards,

 

Patrick


  • Mike in Rancho likes this

#47 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 880
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 10 May 2021 - 06:29 PM

I have StarTools installed now but there are so many parameters. Please post your Startools log if you would be so kind.

 

Regards,

 

Patrick

No problemo.  Give me a few though.  Once I can take a little break I'll delete the mask codes and also try to add a little commentary so it is more useful to you than just replicating settings.  Ivo is best at that of course, but I'll see what I can do.

 

Like anything, playing with it repeatedly and trying different stuff is the best thing to do, at least once you have a basic flow down.  I'm still uncovering "new" stuff myself.

 

Like my D5300 - it didn't matter how many times I read the Nikon reference manual or even "D5300 for Dummies," I didn't really get it until I just started doing it.


  • psugrue likes this

#48 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 186
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 10 May 2021 - 06:58 PM

I can also post the relevant portions of my log, too. Generally, work from the top of the functions to the bottom, with the only exception being that after the wipe module you need to do either autodev or film dev again, then proceed from Wipe, forward. Don't be too afraid of the background seeming a little bright and noisy, it generally works out.
  • psugrue likes this

#49 psugrue

psugrue

    Vostok 1

  • -----
  • topic starter
  • Posts: 144
  • Joined: 11 Nov 2020
  • Loc: Boise Idaho

Posted 10 May 2021 - 07:24 PM

No problemo.  Give me a few though.  Once I can take a little break I'll delete the mask codes and also try to add a little commentary so it is more useful to you than just replicating settings.  Ivo is best at that of course, but I'll see what I can do.

 

Like anything, playing with it repeatedly and trying different stuff is the best thing to do, at least once you have a basic flow down.  I'm still uncovering "new" stuff myself.

 

Like my D5300 - it didn't matter how many times I read the Nikon reference manual or even "D5300 for Dummies," I didn't really get it until I just started doing it.

Thank you Kind Sir!


  • Mike in Rancho likes this

#50 copper280z

copper280z

    Vostok 1

  • -----
  • Posts: 186
  • Joined: 01 Mar 2019
  • Loc: Rochester, NY

Posted 10 May 2021 - 07:28 PM

Here's my ST log, anywhere there's a mask set, I just used the recommended mask type.

 

I think if I were to reprocess, I'd do the 2nd autodev a little differently and allow the background to be a little brighter.

 

Attached File  M81 ST Log.txt   187.06KB   15 downloads


Edited by copper280z, 10 May 2021 - 07:29 PM.

  • Mike in Rancho and psugrue like 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





Also tagged with one or more of these keywords: Astrophotography, Software



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics