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

5 Hours on M81 (Bode's Galaxy)

  • Please log in to reply
21 replies to this topic

#1 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 25 February 2020 - 10:29 AM

I'm up to 5 hours on M81, and wanted to share where I am so far for feedback and encouragement.

 

T5i/700D unmodded

WO Z61 w/Flat61

SkyGuider Pro

Bortle 5/SQM 19.61 (lightpollutionmap.info)

 

Images taken on two nights

 

45 secs x 200 @ ISO1600

60 secs x 150 @ ISO 1600

20 Darks for each session

50 Bias for each session

Flats are driving me nuts! smile.gif  ....so, none

 

Stacked in DSS, processed in PI

 

I am liking where this is headed.  I think that perhaps another night or two will be as far as I want to go with the LP in my backyard.  Thanks for taking a look, any C/C is welcome.

 

Oh, and in case anyone wonders why the crop is a bit tight, I had some pretty crazy stacking artifacts at the edges when I put together the two nights (didn't match the framing as well as I should have), so had to crop it pretty hard.

 

Rich

 

49584066907_9dd1d1ed25_o.jpg


  • elmiko, nemo129, WarmWeatherGuy and 10 others like this

#2 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 25 February 2020 - 01:01 PM

Thought the original I posted was a bit too blue, so pushed the red up a bit.  I like it better....

 

49584676857_2136096696_o.jpg


  • 42itous1 likes this

#3 AstroPics

AstroPics

    Ranger 4

  • *****
  • Posts: 348
  • Joined: 11 Jan 2017
  • Loc: Atlanta, GA

Posted 25 February 2020 - 03:55 PM

First off, really nice image! A few comments/observations if you are interested...

 

I actually think your colors in the first one are better, maybe better calibrated. You are certainly pulling out the spiral arms nicely.

 

I'm curious, are your performing any sharpening in post-processing on your image? It feels like things could be a little tighter but I haven't imaged this object with a DSLR and my AT65EDQ for a long time. Hard to compare the sharpness of your image with my memory of it.

 

The black point seems a little high. This is really subjective though and I think everyone has a different preference.

 

The framing is pretty good from my perspective. It highlights the three galaxies nicely.

 

All that being said, I like what you have done so far!



#4 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 25 February 2020 - 07:18 PM

First off, really nice image! A few comments/observations if you are interested...

 

I actually think your colors in the first one are better, maybe better calibrated. You are certainly pulling out the spiral arms nicely.

 

I'm curious, are your performing any sharpening in post-processing on your image? It feels like things could be a little tighter but I haven't imaged this object with a DSLR and my AT65EDQ for a long time. Hard to compare the sharpness of your image with my memory of it.

 

The black point seems a little high. This is really subjective though and I think everyone has a different preference.

 

The framing is pretty good from my perspective. It highlights the three galaxies nicely.

 

All that being said, I like what you have done so far!

Thanks for the comments!  All of my images right now suffer from some red and blue gradients/artifacts, and the only way I've been able to tame them is by pushing the black point further than is ideal.

 

After my linear processing, I'm doing Local Histogram Equalization, and Unsharp Mask (the latter being the only real sharpening tool I've used.)  I'm open to suggestions on where I could spend a little bit of time.

 

I'm sure more time on the target will help, but, I'm not naive enough to believe that it's not my processing that needs more work! :)  A darker location would be a bonus, but my plan is to just hone my basic capture and processing skills from the backyard before I venture off to a dark(er) site.

 

Thanks again for popping in to comment.  Always good to have more eyes looking at it, especially after a good processing session! ;)

 

Rich



#5 Maged

Maged

    Explorer 1

  • -----
  • Posts: 81
  • Joined: 23 Feb 2015

Posted 26 February 2020 - 06:07 AM

I really like this picture. It's great that the skyguider pro can do. I just got mine and learning my way with it. My only issue with it is that the polar alignment is messed up when I point the scope at the target so I have to re-perform polar alignment again with each new target. Was wondering if you have same issue?



#6 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 26 February 2020 - 09:24 AM

I really like this picture. It's great that the skyguider pro can do. I just got mine and learning my way with it. My only issue with it is that the polar alignment is messed up when I point the scope at the target so I have to re-perform polar alignment again with each new target. Was wondering if you have same issue?

Thanks, I'm very impressed with what I can achieve with this simple mount, and it's been a great platform thus far to start my AP journey.

 

I'm sorry you are having an issue with the Sky Guider Pro and maintaining your PA.  I don't have that issue.  My process is this.  While it's still light out, I set up and level my tripod.  Attach the SGP base, using an app on my phone to ensure that it is roughly pointed toward Polaris.  Remove front polar scope cap, and attach the main SGP unit to the base.  Add counterweight bar, then counterweight.  Load the telescope on top of the unit, then balance the scope in DEC.  Once it is dark enough to see Polaris, I go through the alignment process, using the iOptron PA app on my iPhone to verify where Polaris should be placed in the circle in the scope (in my experience, it is very important to do the PA with all of the gear on the scope as that weight will make the tripod "settle" and throw off PA if you add it after.)  Loosen the DEC clutch, and the RA set screw, "slew" the scope to a bright star for focusing, all while making sure that I don't move the tripod at all.  This process should maintain your PA assuming that you have not moved any of the base components of the set up (tripod, alt/az base and motor mount.)

 

If you are not moving any of these base components after doing polar alignment, you may need to verify that your polar scope is properly aligned with the axis of your motor mount.  I'm pretty sure there's a document somewhere from iOptron on that procedure.  Good luck!

 

Rich



#7 Madratter

Madratter

    Voyager 1

  • *****
  • Posts: 11,975
  • Joined: 14 Jan 2013

Posted 26 February 2020 - 09:39 AM

Thanks for the comments!  All of my images right now suffer from some red and blue gradients/artifacts, and the only way I've been able to tame them is by pushing the black point further than is ideal.

 

I'm going to give you my take on this. Feel free to throw it over your left shoulder.

 

In my view, it is a mistake to mask problems by doing things like pushing up the black point. It inevitably comes at the expense of information that could originally be seen but is now not visible.

 

I would far rather look at an image and see all that it can be, warts and all, than look at an image that has had makeup applied that ends up hiding the flaws but also ends up hiding much of interest as well.

 

That aside, I also thought the original was better than the redo. In particular the color was better.



#8 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 26 February 2020 - 10:43 AM

I'm going to give you my take on this. Feel free to throw it over your left shoulder.

 

In my view, it is a mistake to mask problems by doing things like pushing up the black point. It inevitably comes at the expense of information that could originally be seen but is now not visible.

 

I would far rather look at an image and see all that it can be, warts and all, than look at an image that has had makeup applied that ends up hiding the flaws but also ends up hiding much of interest as well.

 

That aside, I also thought the original was better than the redo. In particular the color was better.

Thanks, Madratter, I appreciate the input for sure!  I haven't developed my own style yet, so taking in all comments will help me see things from angles I haven't considered.  I'll process up a version with more warts, and maybe we can inspect that to get ideas on how I can reduce the imperfections without losing detail/data (I'm still in the mode of following various tutorial steps since I don't have enough experience to have created my own workflow yet.) :)

 

Thanks!

 

Rich



#9 AstroPics

AstroPics

    Ranger 4

  • *****
  • Posts: 348
  • Joined: 11 Jan 2017
  • Loc: Atlanta, GA

Posted 26 February 2020 - 12:47 PM

Thanks for the comments!  All of my images right now suffer from some red and blue gradients/artifacts, and the only way I've been able to tame them is by pushing the black point further than is ideal.

 

After my linear processing, I'm doing Local Histogram Equalization, and Unsharp Mask (the latter being the only real sharpening tool I've used.)  I'm open to suggestions on where I could spend a little bit of time.

 

I'm sure more time on the target will help, but, I'm not naive enough to believe that it's not my processing that needs more work! smile.gif  A darker location would be a bonus, but my plan is to just hone my basic capture and processing skills from the backyard before I venture off to a dark(er) site.

 

Thanks again for popping in to comment.  Always good to have more eyes looking at it, especially after a good processing session! wink.gif

 

Rich

Rich,

 

You might want to look into PixInsight's Deconvolution process during linear processing. I find if this is properly performed, it can bring out detail in galaxy structure and it is also good at tightening up stars. However, learning how to tweak the parameters takes some experience and/or experimentation. If you try this, expect to spend some time learning how to optimize your image without adding background noise or creating artifacts around stars. The investment in time is generally worth it.

 

Keller's 'Inside Pixinsight' gives a good rundown on how to perform a Deconvolution workflow on the luminance component of a OSC image. I imagine there are some good tutorials available as well.

 

Kevin



#10 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 26 February 2020 - 01:26 PM

Nice work!! I like the first version best- colors look better and less star bloating. I took the liberty of grabbing your .jpg and doing a light additional stretch to it and can see why you chose to drop the background due to edges in the stacking. I don't see any weird gradients, but there does seem to be some kind of banding going on with the red channel. It doesn't look like walking noise to me, maybe it's the Canon Banding issue? If it is, it doesn't look like the banding issue that my 5dmkii has. I wonder if they would attenuate out with flats? 

In either case, there is more good data buried in there that you might be able to coax out! 

stretch.JPG



#11 Maged

Maged

    Explorer 1

  • -----
  • Posts: 81
  • Joined: 23 Feb 2015

Posted 26 February 2020 - 01:55 PM

Nice work!! I like the first version best- colors look better and less star bloating. I took the liberty of grabbing your .jpg and doing a light additional stretch to it and can see why you chose to drop the background due to edges in the stacking. I don't see any weird gradients, but there does seem to be some kind of banding going on with the red channel. It doesn't look like walking noise to me, maybe it's the Canon Banding issue? If it is, it doesn't look like the banding issue that my 5dmkii has. I wonder if they would attenuate out with flats? 

In either case, there is more good data buried in there that you might be able to coax out! 

attachicon.gifstretch.JPG

I confirm the red channel banding, I have canon 70D and my photos has the same issue. The flats decrease the issue but it is still noticeable. I am a noob in processing images, maybe there can be something to do about the banding that I don't know!  



#12 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 26 February 2020 - 03:13 PM

Rich,

 

You might want to look into PixInsight's Deconvolution process during linear processing. I find if this is properly performed, it can bring out detail in galaxy structure and it is also good at tightening up stars. However, learning how to tweak the parameters takes some experience and/or experimentation. If you try this, expect to spend some time learning how to optimize your image without adding background noise or creating artifacts around stars. The investment in time is generally worth it.

 

Keller's 'Inside Pixinsight' gives a good rundown on how to perform a Deconvolution workflow on the luminance component of a OSC image. I imagine there are some good tutorials available as well.

 

Kevin

Thanks, Kevin, I'll look into adding some Deconvolution to my linear processing.  I've got Keller's book sitting next to me, just getting started with it. :)

 

Nice work!! I like the first version best- colors look better and less star bloating. I took the liberty of grabbing your .jpg and doing a light additional stretch to it and can see why you chose to drop the background due to edges in the stacking. I don't see any weird gradients, but there does seem to be some kind of banding going on with the red channel. It doesn't look like walking noise to me, maybe it's the Canon Banding issue? If it is, it doesn't look like the banding issue that my 5dmkii has. I wonder if they would attenuate out with flats? 

In either case, there is more good data buried in there that you might be able to coax out! 

attachicon.gifstretch.JPG

Thanks, Zakry, for taking a run at it, even with just a downsized .jpg!!! :)  I have this "banding" in every image, so I'll need to focus some effort on how to minimize/remove it since I'm not planning to upgrade my imaging train anytime soon.  Again, really appreciate you having a go at the data!  That's one of the cool things about this hobby is that you can always go back to "old" data when you have new processing tricks!

 

I confirm the red channel banding, I have canon 70D and my photos has the same issue. The flats decrease the issue but it is still noticeable. I am a noob in processing images, maybe there can be something to do about the banding that I don't know!  

Thanks, Maged.  Good to have a second opinion on the source, even if I still don't know how to get rid of it! :)

 

Rich



#13 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 26 February 2020 - 03:27 PM

If you'd like to pile your work into a dropbox folder and share the link, I'd be happy to take a crack at it. PixInsight does have a Canon Banding removal script!

#14 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 26 February 2020 - 05:17 PM

If you'd like to pile your work into a dropbox folder and share the link, I'd be happy to take a crack at it. PixInsight does have a Canon Banding removal script!

Zakry, two-night, cropped stack in .xsif format is here:  https://www.dropbox...._CROP.xisf?dl=0

 

Would love to see what you come up with (and how.) :)

 

Thanks!

 

Rich


  • zakry3323 likes this

#15 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 26 February 2020 - 05:21 PM

Zakry, two-night, cropped stack in .xsif format is here:  https://www.dropbox...._CROP.xisf?dl=0

 

Would love to see what you come up with (and how.) smile.gif

 

Thanks!

 

Rich

Sounds like fun, I'll check it out this evening when the kiddo's in bed :)


  • rgreenberg2000 likes this

#16 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 26 February 2020 - 09:51 PM

Here's what I came up with after messing around for a bit: 

 

M81_02262020_CROP_clone1.jpg

I feel like deconvolution and noise reduction kind of ended up balancing each other out :)

 

The banding is still an issue, though it is attenuated pretty well with the PixInsight script. The unfortunate thing is that it also cut out much of the red color in the galaxies. This could maybe be fixed with masking over the galaxies using a range selection mask before running the script. I also ended up blowing out most of the star color, but they could be added back by doing a separate RGB stretch with just the stars in mind, and then masking out and replacing them in the final image. Flats would have also helped a bunch :)

 

 


  • rgreenberg2000 likes this

#17 vidrazor

vidrazor

    Ranger 4

  • -----
  • Posts: 357
  • Joined: 31 Oct 2017

Posted 26 February 2020 - 10:00 PM

I'm going to give you my take on this. Feel free to throw it over your left shoulder.

 

In my view, it is a mistake to mask problems by doing things like pushing up the black point. It inevitably comes at the expense of information that could originally be seen but is now not visible.

 

I would far rather look at an image and see all that it can be, warts and all, than look at an image that has had makeup applied that ends up hiding the flaws but also ends up hiding much of interest as well.

 

That aside, I also thought the original was better than the redo. In particular the color was better.

Although I've learned not to black out as much, I prefer to sacrifice subtle data for a richer black. It's all a personal preference of course. I also know how to extract luma masks that help me preserve subtle data even against blacker skies.



#18 vidrazor

vidrazor

    Ranger 4

  • -----
  • Posts: 357
  • Joined: 31 Oct 2017

Posted 26 February 2020 - 10:44 PM

Here's what I came up with after messing around for a bit: 

 

attachicon.gifM81_02262020_CROP_clone1.jpg

I feel like deconvolution and noise reduction kind of ended up balancing each other out smile.gif

 

The banding is still an issue, though it is attenuated pretty well with the PixInsight script. The unfortunate thing is that it also cut out much of the red color in the galaxies. This could maybe be fixed with masking over the galaxies using a range selection mask before running the script. I also ended up blowing out most of the star color, but they could be added back by doing a separate RGB stretch with just the stars in mind, and then masking out and replacing them in the final image. Flats would have also helped a bunch smile.gif

I decided to hack your render up through luminosity masks to preserve the highlights and dust data while bringing the sky much darker. I also used various processes to open up edges. This makes for greater contrast and some saturation. Have a look:

Attached Thumbnails

  • hack test.jpg

  • rgreenberg2000 and zakry3323 like this

#19 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 26 February 2020 - 11:03 PM

Zakry and Vidrazor, thanks a ton for putting your spin on my data.  I think that's one of the best learning tools is to see what can be done with the same data, and I really appreciate the insight into the processes you applied.  There's definitely a stylistic portion to processing that can be very specific to one's tastes, but just getting introduced to the impact of processes I haven't used is incredibly helpful.

 

Thanks, to both of you again!

 

Rich


  • zakry3323 likes this

#20 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 27 February 2020 - 01:07 PM

Sure thing! Hey, what issue are you having with flats? When it comes to getting better signal to noise, you wouldn't need to drop the background so much if you were to calibrate with flats- not only will they take care of dust motes and edge vignetting, but they'll also even out the illumination in the background. There's a pretty significant dropoff throughout the frame, check it out: 

 

RangeMask.JPG



#21 rgreenberg2000

rgreenberg2000

    Explorer 1

  • -----
  • topic starter
  • Posts: 58
  • Joined: 17 Jul 2012
  • Loc: San Mateo County

Posted 27 February 2020 - 01:22 PM

Sure thing! Hey, what issue are you having with flats? When it comes to getting better signal to noise, you wouldn't need to drop the background so much if you were to calibrate with flats- not only will they take care of dust motes and edge vignetting, but they'll also even out the illumination in the background. There's a pretty significant dropoff throughout the frame, check it out: 

 

attachicon.gifRangeMask.JPG

I was trying to take flats using a cheap LED screen, and when I used them to calibrate, they were making my images worse.  I don't have much experience with taking flats, so I didn't really have a good path forward to analyze the flats I had taken (hurriedly, without much thought process), so I just hadn't gotten to it again yet.  My plan is to use the N.I.N.A. flat wizard the next time I have an imaging session, and will probably just use the t-shirt/bright sky method.  I have been really focusing on some of the basics to try and capture the best data I can so that my process for setup (level, PA, balance, etc.) is really good, and somewhat second nature.  Flats are my next process to get nailed down, and then will start to work on guiding.

 

I know that I will miss out on the opportunity to use flats to improve my data, but I needed to work on all of the components in a more serial manner to not drive myself crazy.  I've been marginally successful with that. :)

 

I'll get there, and the skies aren't going anywhere (though they may get more crowded....) ;)

 

Rich


  • zakry3323 likes this

#22 zakry3323

zakry3323

    Apollo

  • *****
  • Posts: 1,414
  • Joined: 11 Apr 2016
  • Loc: Pittsburgh

Posted 27 February 2020 - 01:46 PM

I understand completely, and it sounds like you have a good plan to continue to learn and move forward. Flats can be challenging to nail down successfully. I haven't used NINA, but I'm sure it will do a fine job. The SGP Wizard probably works in a similar way. Lots of Flats topics here on CN, you're definitely not alone in having issues. Sky flats have worked very well for me, but the timing has to be just right, and illumination changes as the sky gets darker/brighter depending on whether you're taking them in the evening or morning, so that's something to be wary of- If you take say, 5sec flats 30x, the ADU value will change a little from your first exposure to your last. For OSC I think that's fine- for LRGB it gets a little inconvenient for me using the wizard for sky flats because the ADU measurements taken at the beginning of the session tend to be invalid by the time I'm shooting flats for my last filter. 

You can negate that issue by shooting a blue sky, or a overcast sky, with a diffuser (like a t-shirt). That way illumination isn't really changing much. The issue for me there is that weather can be very unpredictable and I may not have the opportunity to take proper flats right before or after my imaging session with that method. Also shooting blue-sky requires a LOT of diffusion to keep exposures in the "few seconds" length- for me that's using a t-shirt and several layers of opaque acrylic. 

 

There are certainly more elegant solutions, especially if you have an observatory to hang an adjustable illuminated panel, but that's unfortunately outside of my experience (and likely will be for a very long time) smile.gif


Edited by zakry3323, 27 February 2020 - 01:53 PM.

  • rgreenberg2000 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