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

M101 (Pinwheel Galaxy) in LRGB with C11 - C&C Welcome

astrophotography cassegrain ccd Celestron dso imaging SCT
  • Please log in to reply
12 replies to this topic

#1 BenKolt

BenKolt

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1967
  • Joined: 13 Mar 2013

Posted 21 May 2019 - 11:42 PM

Greetings!

 

At last I have called it quits (at least for now) processing lots of LRGB data I managed to capture of M101.  I'm mostly pleased with the result and am even more pleased to actually produce something rather than dream wistfully of summer.

 

I include my very long description from AstroBin below, but for those who don't care to read yet again the physical description of the Pinwheel Galaxy nor to slog through my scintillating musings, here is the image.

 

As always, your comments and suggestions are most welcome!  Thank you for your attention.

 

Best Regards,

Ben

 

 

Higher Resolution Links:

M101_LRGB2_089-090_0300s_20190425_-25C.v003A_LRGB-1.jpg

 

 

From the Description on the AstroBin page ...

 

Physical Details

 

M101 (the Pinwheel Galaxy) in Ursa Major (UMa) is an iconic face-on grand design spiral galaxy.  With an apparent size of 29' x 27', it filled my FOV using the QSI 683 with the Celestron 11" EDGE, even with the Celestron 0.7x focal reducer.  I did manage to encompass nearly all of it including the prominent arm splayed off to the left in this image as well as two of the fainter arms protruding to the right.

 

M101 is 21 Mly (6 Mpc) distant with diameter 170 kly with about 1 trillion stars.  Interestingly, our Milky Way is larger at 258 kly diameter but with only about half that number of stars.  There are many HII star forming regions, several of which can clearly be seen in this image in red.  Several are large enough to have received NGC designations, as seen in my annotated Version D.

 

One of M101's distinct features is its asymmetry having resulted from interaction with nearby galaxies, which can't be seen in this image at this scale.  This causes strong star forming interaction in the spiral arms, denoted by a mostly blue hue that this image has captured.

 

 

Acquisition and Processing Details

 

I had a string of welcome, mostly clear nights at the end of April through early May.  I devoted several nights to accumulating more frames than usual of M101 in L and R,G,B.  With more data than usual, I could be both selective in using frames for the combined channels as well as to include more integration time and draw out better detail.  This was also quite necessary since many of the nights had low transparency, however thankfully average to above-average seeing.

 

This is my first posted image having used the Mure Denoise script in PixInsight.  I'm quite pleased with the outcome and like the way this process utilizes the camera's measured characteristics to apply noise reduction.  The result is quite intuitive and pleasing.

 

With 10h of L at 1x1 I spent quite a bit of time working on the Deconvolution process.  I am still somewhat bothered thinking that some of the brighter regions of galaxy have been over-deblurred to the point where they look like little pinpoint stars, but others have informed me that these may actually be stars after all.  I'm not sure about that yet, but I'm also done messing with it for now, so here it is.  (Examining the Hubble image - see below - of M101, however, I'm now thinking that these pinpoints are real and I haven't created detail that isn't there after all.)

 

The coloring doesn't quite look right to me.  The initial result showed too much green, particularly on the right side of the galaxy in its more nebulous regions, so I redid the RGB processing to apply more aggressive green removal with SCNR during the linear phase.  This ended up with a more pleasing result, however I note that the histogram shows a distinct suppression of the green to the left.  I prefer to see the colors lined up better around the background peak.  In addition to this I can still see some mottled red and blue here and there, so even with quite a bit of integration time my colors aren't quite what I wanted.  Perhaps with a future redo I will select fewer frames and see if the color quality improves.

 

Finally, I am not completely satisfied with the clipping of the stars.  I determined an optimal exposure time of 300s for all channels, and in retrospect I now wish I had backed this off a little bit to not saturate the stars so much.  My next imaging will be with my new FLI ML16200 camera, which I already know will tolerate much shorter exposure times.

 

 

Conclusion

 

Overall I'm still pleased with this, particularly as the image marks a fun, unexpected span of many nights of imaging around the New Moon where I could capture lots of frames of one of my favorites.

 

Lastly, I want to remind the reader of the amazing composite of Hubble images of M101 that can be examined extremely closely.  I don't wish to detract from my hard work, but the Hubble image is truly beautiful and intriguing to explore.  It has been fun for me to compare my image with Hubble's and am pleased to note that many of the background galaxies around and behind M101 can actually be spotted in my image as well, albeit a tad more blurry.

 

Thank you for your attention, and as usual, your comments and suggestions are most welcome!

 

 


  • astrokwang2, Alterf, Joe F Gafford and 14 others like this

#2 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3995
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 21 May 2019 - 11:57 PM

Beautifully done. Looks like youve judged the sharpening and denoising pretty darn well.



#3 44maurer

44maurer

    Viking 1

  • *****
  • Posts: 585
  • Joined: 12 Jun 2016
  • Loc: Los Angeles, CA

Posted 22 May 2019 - 12:27 AM

Another great image. You managed to get a few nights of clear skies.

Thanks for the additional details.

#4 astrovienna

astrovienna

    Mercury-Atlas

  • *****
  • Posts: 2771
  • Joined: 04 Dec 2006
  • Loc: The NoVa White Zone

Posted 22 May 2019 - 09:59 AM

Ben, that's an excellent result.  One of the nicest M101s I've seen in a long time.  I agree with the others on your deconvolution result - you're getting the best possible outcome.  Even if not everything that becomes pointlike is really a star, I don't think anyone (except the author, of course!) would be aware of that.  And like a poster on your other thread, I meant to point out that my understanding from the author of Mure Denoise is that it should be run before anything else, such as DBE.  But your result worked out fine.

 

One very small issue is the color fringes on stars at the edges of the FOV.  Do you run into that often with the Edge scope?  It's a constant headache with my non-Edge C11, but I thought the Edge optics had solved this.  The way I deal with this is in Photoshop.  I create a color layer, and then use Filter>Blur>Radial Blur to "spin" each star.  That blurs the color fringes down to the dominant color.  A kludge, but it works for me.  Anyway, just a very minor suggestion on a great image.

 

Kevin



#5 BenKolt

BenKolt

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1967
  • Joined: 13 Mar 2013

Posted 22 May 2019 - 10:44 AM

Beautifully done. Looks like youve judged the sharpening and denoising pretty darn well.

Thank you!  I've been paying more attention to those steps rather than just applying the processes in a robotic fashion.

 

Another great image. You managed to get a few nights of clear skies.

Thanks for the additional details.

Thanks, Brian!  This was a rare stretch of nights for me this time of year, and for them to coincide closely with the New Moon was even more of a rarity.

 

Ben, that's an excellent result.  One of the nicest M101s I've seen in a long time.  I agree with the others on your deconvolution result - you're getting the best possible outcome.  Even if not everything that becomes pointlike is really a star, I don't think anyone (except the author, of course!) would be aware of that.  And like a poster on your other thread, I meant to point out that my understanding from the author of Mure Denoise is that it should be run before anything else, such as DBE.  But your result worked out fine.

 

One very small issue is the color fringes on stars at the edges of the FOV.  Do you run into that often with the Edge scope?  It's a constant headache with my non-Edge C11, but I thought the Edge optics had solved this.  The way I deal with this is in Photoshop.  I create a color layer, and then use Filter>Blur>Radial Blur to "spin" each star.  That blurs the color fringes down to the dominant color.  A kludge, but it works for me.  Anyway, just a very minor suggestion on a great image.

 

Kevin

Thanks, Kevin!  Having looked at the huge Hubble mosaic of M101, I can now see that many of those pinpoint things are in that image as well, so at least I'm in good company.

 

Interestingly, I found that MureDenoise (MD) did not work well on some of my channels (particularly B) unless I first applied a gradient removal.  Once I flattened out the background, MD seemed to work well for me, so I am supposing that as long as I was only removing very low frequency spatial background features there was no ill effect.  I'm still unclear, however, why MD messed up so badly on the combined image prior to background removal, but only on some of the channels.

 

Yes, you caught the color fringes around the edges of the FOV!  For this image I used the Celestron 0.7x Focal Reducer (FR), which I really don't like.  With the QSI 683 camera, I had to use the FR if I wanted to fit most of the galaxy in the FOV.  But this is the primary source of my fringing trouble.  I have noticed that when I don't use the FR the effect is much minimized or not even apparent.  It's possible that some fine tweaking of back focus spacing could help, but I am not the only one to complain about this FR!  I had hoped that StarAlignment with distortions settings checked would have helped, but the stars are contorted enough with respect to one another across the color channels that I still got the distortions.  Next time I'll look into what you suggest to mitigate the problem.  Thanks!

 

Ben



#6 astrovienna

astrovienna

    Mercury-Atlas

  • *****
  • Posts: 2771
  • Joined: 04 Dec 2006
  • Loc: The NoVa White Zone

Posted 22 May 2019 - 02:29 PM

Thank you!  I've been paying more attention to those steps rather than just applying the processes in a robotic fashion.

 

Thanks, Brian!  This was a rare stretch of nights for me this time of year, and for them to coincide closely with the New Moon was even more of a rarity.

 

Thanks, Kevin!  Having looked at the huge Hubble mosaic of M101, I can now see that many of those pinpoint things are in that image as well, so at least I'm in good company.

 

Interestingly, I found that MureDenoise (MD) did not work well on some of my channels (particularly B) unless I first applied a gradient removal.  Once I flattened out the background, MD seemed to work well for me, so I am supposing that as long as I was only removing very low frequency spatial background features there was no ill effect.  I'm still unclear, however, why MD messed up so badly on the combined image prior to background removal, but only on some of the channels.

 

Yes, you caught the color fringes around the edges of the FOV!  For this image I used the Celestron 0.7x Focal Reducer (FR), which I really don't like.  With the QSI 683 camera, I had to use the FR if I wanted to fit most of the galaxy in the FOV.  But this is the primary source of my fringing trouble.  I have noticed that when I don't use the FR the effect is much minimized or not even apparent.  It's possible that some fine tweaking of back focus spacing could help, but I am not the only one to complain about this FR!  I had hoped that StarAlignment with distortions settings checked would have helped, but the stars are contorted enough with respect to one another across the color channels that I still got the distortions.  Next time I'll look into what you suggest to mitigate the problem.  Thanks!

 

Ben

Interesting about Mure Denoise.  When you say it didn't work well, do you mean it was too aggressive, or not having any effect?  I've sometimes found it has no effect on my luminance images, which I've never understood.  It usually works like a charm, though.

 

Kevin



#7 BenKolt

BenKolt

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1967
  • Joined: 13 Mar 2013

Posted 22 May 2019 - 06:21 PM

Interesting about Mure Denoise.  When you say it didn't work well, do you mean it was too aggressive, or not having any effect?  I've sometimes found it has no effect on my luminance images, which I've never understood.  It usually works like a charm, though.

 

Kevin

It was way too aggressive with the blue channel image.  It had a notable background gradient, and my blue channel is usually my weakest anyway with this camera.  What I figure is that the algorithm determined much of the image was background in need of noise reduction.  Unfortunately, this affected a good deal of the signal detail as well.  After background gradient removal, the algorithm worked just fine on the whole image.

 

I have since become concerned that my flat panel may not be as flat as I think and could be the cause for some of these issues. (I've had my panel for several years now and it's gotten lots of use.  They don't come with an expiration date, but I can only assume there is one, particularly concerning uniformity.)  I've been performing some tests on it, and my conclusion thus far is that that panel has a definite nonuniform brightness primarily along one axis.  One quick measurement showed as much as 6% along this axis, and that's not acceptable in my book.  Unfortunately I was unable to take sky flats during the time of my imaging, otherwise I'd use those, but sky flats are often problematic for me because I'm either unavailable, there are variable clouds, etc.

 

A new panel may be on the horizon for me.  More testing tonight.

 

Ben



#8 stargzr66207

stargzr66207

    Ranger 4

  • -----
  • Posts: 367
  • Joined: 28 Aug 2014
  • Loc: Kansas, USA

Posted 22 May 2019 - 06:44 PM

Ben,
That is well done! Congratulations on your entire process.
Ron Abbott

#9 elmiko

elmiko

    Gemini

  • -----
  • Posts: 3325
  • Joined: 27 Jul 2009
  • Loc: Arizona

Posted 22 May 2019 - 08:09 PM

Ben, you did a wonderful job on this target. Your processing was very well done. I like the color also.

Mike



#10 lucam

lucam

    Viking 1

  • -----
  • Posts: 599
  • Joined: 09 Oct 2017
  • Loc: Upstate NY, USA

Posted 22 May 2019 - 08:16 PM

Ben,

 

That's a fantastic result! A few of thoughts:

 

- I haven't had a chance to view your image on my calibrated monitor at home but as you mentioned there appears to be a subtle color gradient (purple to green). You really have to look to find it. 

- I think you can push the saturation a bit especially in the core region. The image is subtle and the detail is incredible but I think you can make it pop a bit more without sacrificing resolution or color fidelity. 

- Your HII regions are a bit pale. I don't think you used Ha data but the image would really gain from a small Ha bump in the red channel. I would not touch the luminance. If you want to experiment, I have about six hours of Ha data on my Pinwheel at 0.68 arcsec/px, which I would be happy to share with you. 

 

Regarding MureDenoise, have you tried to use flats in the tool? With such a narrow field of view, it's unlikely you have much of a sky gradient. At NEAIC, Adam Block mentioned that he almost always applies DBE as a division because it usually takes care of leftover flatfield correction more than sky gradient. Obviously that's not the case for wide fields of view, but in your case I suspect it may be, especially when you say your panel may not be particularly homogeneous. That may be the reason why MD works more uniformly after you remove the "gradient". 

 

Again, I am nitpicking just because I have been obsessing about this DSO recently but as others have said, it's the best, most detailed M101 I have seen in a while. 



#11 flyingcougar

flyingcougar

    Apollo

  • *****
  • Moderators
  • Posts: 1428
  • Joined: 11 Apr 2008
  • Loc: New Mexico

Posted 22 May 2019 - 08:26 PM

Hi Ben, that's a fantastic image and the processing is beyond anything I could manage. I will agree with Luca regarding the Ha data on this subject, it would really make it stand out IMO.

 

Congrats on a great image!



#12 DaveB

DaveB

    Apollo

  • *****
  • Posts: 1360
  • Joined: 21 Nov 2007
  • Loc: New England

Posted 22 May 2019 - 08:47 PM

That's really an amazing image! bow.gif   Very crisp details without looking overly sharpened or otherwise processed. That is the type of look that I strive for in my images. One of these days, I'll achieve something close to it... 



#13 BenKolt

BenKolt

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1967
  • Joined: 13 Mar 2013

Posted 23 May 2019 - 12:10 AM

Ben,
That is well done! Congratulations on your entire process.
Ron Abbott

Thank you, Ron! Lately I've been spending a whole lot more time with the steps rather than to simply run the processes in a robotic fashion as before.

 

 

Ben, you did a wonderful job on this target. Your processing was very well done. I like the color also.

Mike

And thank you, Mike!  Colors are a subjective choice, and I usually tend to go more subtle.

 

 

Ben,

 

That's a fantastic result! A few of thoughts:

 

- I haven't had a chance to view your image on my calibrated monitor at home but as you mentioned there appears to be a subtle color gradient (purple to green). You really have to look to find it. 

- I think you can push the saturation a bit especially in the core region. The image is subtle and the detail is incredible but I think you can make it pop a bit more without sacrificing resolution or color fidelity. 

- Your HII regions are a bit pale. I don't think you used Ha data but the image would really gain from a small Ha bump in the red channel. I would not touch the luminance. If you want to experiment, I have about six hours of Ha data on my Pinwheel at 0.68 arcsec/px, which I would be happy to share with you. 

 

Regarding MureDenoise, have you tried to use flats in the tool? With such a narrow field of view, it's unlikely you have much of a sky gradient. At NEAIC, Adam Block mentioned that he almost always applies DBE as a division because it usually takes care of leftover flatfield correction more than sky gradient. Obviously that's not the case for wide fields of view, but in your case I suspect it may be, especially when you say your panel may not be particularly homogeneous. That may be the reason why MD works more uniformly after you remove the "gradient". 

 

Again, I am nitpicking just because I have been obsessing about this DSO recently but as others have said, it's the best, most detailed M101 I have seen in a while. 

Thanks, Luca, especially for the suggestions.  No, I did not use Ha data here, just straightforward LRGB, so the HII regions are more subtle as you noticed.

 

There was actually quite a gradient, but over the past couple of days I've determined that my flat panel does indeed have a notable nonuniformity, and I'm afraid now that I've been inducing my own gradients beyond the normal sky background!  That's a real bummer, of course.  Perhaps some of my color issues stem from this issue.  A new flat panel may well be in my future ...

 

I appreciate your kind offer of letting me play with your Ha data, but I'm really not going to have the time to work with it any time soon due to all my projects.  I'll keep your offer in mind for a future time if I wish to return to this data and jazz it up.

 

I'll look for Adam Block's method with DBE and see if that helps me in the future.  Thanks for the suggestion!

 

 

Hi Ben, that's a fantastic image and the processing is beyond anything I could manage. I will agree with Luca regarding the Ha data on this subject, it would really make it stand out IMO.

 

Congrats on a great image!

Thank you, Keith!  I'll think about Ha at a later time.  Earlier today I checked to see if M101 would work with my 3nm filter, and indeed it should since it's a fairly close galaxy with a small enough redshift.  It turns out there actually aren't that many galaxies available for such a small bandwidth.

 

 

That's really an amazing image! bow.gif   Very crisp details without looking overly sharpened or otherwise processed. That is the type of look that I strive for in my images. One of these days, I'll achieve something close to it... 

Thanks, Dave!




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, cassegrain, ccd, Celestron, dso, imaging, SCT



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics