CaK from August 7th with ARIES-ERF

this is the main message area for anything solar :)
Post Reply
LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

CaK from August 7th with ARIES-ERF

Post by LTHB » Wed Aug 08, 2018 2:07 pm

Sun is shining even in northern Germany now AND seeing was o.k. yesterday, so another rare opportunity to get the ARIES-ERF out and capture some CaK.

Image

This is the result of two afternoons trying to understand how to use imppg, i.e. deconvolution, to get the best out ot the data. While I am quite pleased with the detail in the plage, I an unsure how much of the grainy pattern/structure in the darker parts are processing artifacts.

Comments and critique welcome! (The scope is a Mewlon 210 with a 2xBarlow and a Lunt CaK filter double-stacked with a PST filter.)

Kind regards,

Frank

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 19677
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: CaK from August 7th with ARIES-ERF

Post by Montana » Wed Aug 08, 2018 2:20 pm

That looks top notch to me, very fine subtle processing maintaining smooth solar features. No noise either, very nice. The super granulation cells are always of mottled appearance, they look correct :bow

Alexandra

User avatar
robert
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 2397
Joined: Fri Oct 14, 2011 1:49 pm
Location: Isle of Skye, Scotland
Contact:

Re: CaK from August 7th with ARIES-ERF

Post by robert » Wed Aug 08, 2018 5:00 pm

Lovely detailed shot
Robert
images and animations http://tinyurl.com/h5bgoso
2018 images https://www.flickr.com/photos/69734017@ ... 1234171714
2017 images http://tinyurl.com/gvk9k3k
ED80. SW200. ES127. Celestron-150mm-PST mod
LS60PT-LS60F-B1200. B600-Cak. PGR-Ch3-IMX265

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3010
Joined: Sun Nov 18, 2012 3:13 pm

Re: CaK from August 7th with ARIES-ERF

Post by Valery » Wed Aug 08, 2018 8:06 pm

Hi Frank,

Good result for the very beginning. Looks underprocessed. Many depends of what was the seeing and how many frames have you stacked. You can send me some unprocessed stacked images and I will take care of them and see how much deeper you can go with processing to extract more and finer details.


Valery
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Thu Aug 09, 2018 1:31 pm

Thanks, Alexandra and Valery!

Valery, I just try to send you a PM with the unprocessed stack from Autostakkert3, but failed, probably because the file is too large (15MB). Can I send it via email? - It would be great if I could see your processing results and learn!

I also just searched a little for high resolution CaK images of comparable resolution/image scale on this forum. I found a few from different people, but not many. While most of them are really beautiful, different people obviously process their CaK images in very different ways and accordingly the results look quite different (i.e. high res CaK pictures, not so much the full disk or quarter disk images) and show quite different structures and textures. Now one may, of course, enjoy the sheer beauty of those pictures - but I also found the differences a little puzzling as I would like to know which structures are, so to speak, depicting real structures of the sun and which are, maybe, artifacts of processing (in particular of my newbie processing...).

I found this one year old thread:

viewtopic.php?f=4&t=22697

very interesting and helpful, in particular the images from the Dutch Open Telescope. If they may be taken as a reference/standard of an accurate portrayal of the sun's features in CaK (but I am not sure if or how far that makes sense), one might try to scale them down (or apply a convolution to model the limitations of our smaller scopes?) to get an idea of how a CaK image of amateur size scopes could/should look if it shows no processing artifacts. Does that make sense?

Regards,

Frank

User avatar
eroel
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 4052
Joined: Mon Dec 19, 2011 10:45 pm
Location: México D.F.

Re: CaK from August 7th with ARIES-ERF

Post by eroel » Thu Aug 09, 2018 2:29 pm

Frank:
Nice clean image but looks a bit unprocessed.
Best regards,
Eric.

User avatar
marktownley
Librarian
Librarian
Posts: 24378
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: CaK from August 7th with ARIES-ERF

Post by marktownley » Thu Aug 09, 2018 2:31 pm

This is an excellent image Frank! Hope you get to take many more :)
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Fri Aug 10, 2018 12:32 am

Thanks, Eric and Mark!

Eric, I can see why the picture looks underprocessed to you (and to Valery, too),while in fact I did quite a lot of processing: It looks slightly blurred, but I could not improve that by much. My second version:

Image

Regards,

Frank

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3010
Joined: Sun Nov 18, 2012 3:13 pm

Re: CaK from August 7th with ARIES-ERF

Post by Valery » Fri Aug 10, 2018 1:59 pm

Hi Frank,

BTW. Your last image looks like my unprocessed images.

You can put the image in the PhotoShop and cut the best ROI and make it smaller in PGR format but still 16 bit.
This will be OK for processing. Yes, send me the best frame via e-mail.

Valery
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3010
Joined: Sun Nov 18, 2012 3:13 pm

Re: CaK from August 7th with ARIES-ERF

Post by Valery » Fri Aug 10, 2018 2:23 pm

LTHB wrote:
Fri Aug 10, 2018 12:32 am
Thanks, Eric and Mark!

Eric, I can see why the picture looks underprocessed to you (and to Valery, too),while in fact I did quite a lot of processing: It looks slightly blurred, but I could not improve that by much. My second version:



Regards,

Frank

Here is your last image re-processed.

1. The filter is somewhat off-band (blue wing)
2. Image is very undersampled (way too small scale).

3. Seeing was good.


Valery
Attachments
FRANK-reprocessed.jpg
FRANK-reprocessed.jpg (923.54 KiB) Viewed 483 times
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Fri Aug 10, 2018 4:15 pm

Hi Valery,

I just sent you a PM with the unprocessed stack of the image in my first post (a tif-file).

After that I just saw your post with the reprocessed image, thanks! That's really interesting and it speaks precisely to the point of my question!

I have used imppg's deconvolution tool for processing and of course I tried out several settings for sigma and the number of iterations. One of my versions looked similar to yours - not quite as clean, but similar: much less blurred and hazy than the version I posted, but I discarded that because I took the "worm-like" look of the structures (I don't know how to describe it more accurately) to be processing artefacts.

My observation was that when you try out different sigma values for deconvolution there is a kind of trade-off between haziness and "worm-like" look of the resulting picture: When you start with the unprocessed stack and raise the sigma value little by little, you lose more and more of the haziness, but the fine surface structures that show up after raising sigma a few steps then soon begin to take on a worm-like shape: the higher sigma, the more worm-likeness and the thicker the worms, so to speak. Needless to say, the worms begin to show up way before the haziness is completely gone.

While I would, of course, have preferred an image without haziness AND without worms, I dedided to halt the deconvolution before the worms appeared and to put up with a measure of haziness (the underprocessed look in Eric's and your view). My reason was that I suppose that I am losing a bit of the fine detail/information in the picture as soon as the worms show up: All the worms seem to me to have (more or less) the same thickness, on the light parts of the surface as well as on the darker parts, while in the hazy version of the image I can distinguish some coarser structures and also thinner, finer structure, in particular in the lighter areas of the plage. If THIS fine detail/structure corresponds to something real on the sun, or if this is precisely (part of) the noise in the stack that SHOULD be removed when the image is processed appropriately, is precisely the point that I'd like to know. (The beauty of the pictures and aesthetic preferences are, of course, another matter.)

The pictures from the DOT shown in the old thread mentioned above and similar pictures might be a way to argue one way or another. If a find a little more time I will try to look into this.

Thanks, Valery, also for your other hints! I will try out a camera with smaller pixels when I get the next chance for decent weather and seeing and hopefully get higher resolution.

I do not quite know what to make of your observation that the filter is somewhat off-band. How do you see this in the picture? To my knowledge the CaK filters (by Lunt as well as Coronado) are a few Angstroms wide, so they should certainly cover the complete CaK line, but I did not check this in any way for my filter. Also the filters are not tilted or heated. Any idea what could get a CaK-filter off-band?

Thanks again and regards,

Frank

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3010
Joined: Sun Nov 18, 2012 3:13 pm

Re: CaK from August 7th with ARIES-ERF

Post by Valery » Sat Aug 11, 2018 5:54 pm

Hello Frank,

I am unable to read the attached file in the PM you sent me.

Please, send it to my e-mail.


Vakery
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Sat Aug 11, 2018 7:26 pm

Hi all,

looking for a standard for correct/appropriate processing of my CaK images I browsed the webpage of the (now mothballed) Dutch Open Telescope (DOT). I found this one:

Image

The DOT is, of course, much bigger than our amateur scopes, so it is not realistic to hope for pictures of this resolution and quality. I wanted to try if the DOT picture can serve as a standard for our amateur imaging anyway if scaled down and blurred somewhat to model the influence of our smaller apertures.

So I downloaded the picture and applied a Gaußian blur with 4px and scaled it down to 33 percent, so that the image scale should be roughly the same as in my own picture:

Image

(On the DOT webpage they explicitly state: "You are welcome to use any DOT material in publications and presentations of any kind [...] We appreciate acknowledgement to "Dutch Open Telescope" (short), or "Dutch Open Telescope, operated at the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofísica de Canarias" (long)", so I think it should be okay in terms of copyright to process and post the image here - acknowledging the DOTs fabulous images, of course!)

I picked the section on the lower left-hand side for closer comparison with a similar looking section from the center of my picture and in my own and in Valery's reprocessed versions:

Image
Detail from DOT image

Image
Detail from my image

Image
Detail from Valery's reprocessed version

I think, if the idea to use the scaled down and blurred version of the DOT image as a standard is reasonable, then, looking at these three cut-outs side by side, it is very obvious that Valery and Erik are right: My version is underprocessed: way too soft and blurry.

On the other hand Valery's version looks a little bit oversharpened to me, with contrast a little too harsh, allthough his version clearly comes much closer to the "DOT look" than mine. In particular it shows (even if a bit oversharpened) the structures in the dark sections of the image that are more or less washed out in my version.

Finally, taking a look at other people's high resolution CaK images I think at least from those I have found on this forum Christian Viladrich's probably come closest to the "DOT look", i.e. the high res-CaK here: viewtopic.php?f=4&t=22556&p=205387#p205387
Maybe he can give us some advice how to improve our processing!

Meanwhile I have, of course, processed my CaK stack once again and tried to get closer to the "DOT look". The result in detail:

Image


and the complete picture:

Image

So this is at the moment my best guess of how to process high res CaK-images if the aim is to bring out as much detail of the real solar structures as our amateur scopes can capture and not to bring in too many processing artefacts.

Comments and critique welcome!

Frank

christian viladrich
Im an EXPERT!
Im an EXPERT!
Posts: 426
Joined: Sun Jun 14, 2015 4:46 pm
Location: France
Contact:

Re: CaK from August 7th with ARIES-ERF

Post by christian viladrich » Wed Aug 22, 2018 7:37 pm

Hello Frank,
If I have a suggestion, this would be to forget about deconvolution and only apply an unsharp mask.
Deconvolution won't replace a lack of resolution in the image.

Here is a recent Ca K image taken with the TOA 150 :
Image

The main structure visible is the reverse granulation.

More images are visible here :
http://astrosurf.com/viladrich/astro/so ... _2018.html
Christian Viladrich
Co-author of "Astronomie Planétaire"
http://www.astroplanetes.com/
Co-author of "Astronomie Solaire"
http://www.astronomiesolaire.com/

User avatar
p_zetner
Almost There...
Almost There...
Posts: 894
Joined: Sun Mar 18, 2012 4:59 pm
Location: Winnipeg, Manitoba, Canada

Re: CaK from August 7th with ARIES-ERF

Post by p_zetner » Thu Aug 23, 2018 9:19 pm

A very informative thread. Thanks for posting this, Frank.
I've also been concerned with overprocessing high resolution CaK with too much deconvolution. The DOT image makes a good reference to help judge this.

Cheers, Peter.

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Sun Aug 26, 2018 10:53 am

Hello Christian,

thanks for the suggestion!

When I apply a (rather aggressive) unsharp mask only (i.e. no deconvolution) the best I get looks like this:

Image

Unfortunately this does not really seem to improve the image: While the difference to my last version (in the post above) is certainly not great, the version using deconvolution seems to me to bring out the details slightly better (allthough not a lot better).

I spent quite a while looking at the impressive array of solar images on your websites (and also on your enviable observatory's website that you linked in another thread). Keeping an open eye for clues to improve my imaging, one thing that struck me was that your stacks are smaller (mostly around 50-60 frames) than mine (mostly around 200 frames) and also smaller than the stacks used for most solar images I have seen on the web. (I also recall Emil Kraaikamp writing somewhere on his website that he prefers stacking a larger number of frames and then, as he puts it, pushing the stack a bit harder.) When I try to stack "only" the 50 or so best frames of my solar videos, Autostakkert's output is somewhat sharper than stacking 200 or 300 frames, but very noisy, and so far I have not found a way to remove that noise (while keeping the sharpness), so I usually stack around 200 frames. Now, one hint I gleaned from your images is that your cameras have much less read noise than mine, so, assuming that the noise is at least in part read noise, one way to improve may be to find a camera with lower read noise in order to work with smaller, less noisy stacks.

One more question, if I may: You remarked that the main structure visible on your CaK images is "reverse granulation". I did not know (and had not heard or read) that before and I find this very interesting. Looking at your images I suppose you are referring to the darker, grey areas and the short "crinkled" lines there. Are these then the boundaries of photospheric granulation cells seen in absorption? (If yes: I wouldn't have guessed that because they look more criss-crossed than what I would expect from granulation cell boundaries. Are there, maybe, images where one can correlate the CaK pattern with white light granulation structures?)

Thanks again and kind regards,

Frank

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3010
Joined: Sun Nov 18, 2012 3:13 pm

Re: CaK from August 7th with ARIES-ERF

Post by Valery » Sun Aug 26, 2018 3:41 pm

Frank,

I see a hidden potential in this your image.

Send me it via e-mail and I will try to extract it's potentiall to a higher degree.

Better a 16 bit and 1000, 1500, 2000 frames stacked.


Valery
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

LTHB
Oh, I get it now!
Oh, I get it now!
Posts: 41
Joined: Thu Mar 03, 2016 7:31 pm

Re: CaK from August 7th with ARIES-ERF

Post by LTHB » Mon Aug 27, 2018 7:38 pm

Hi Valery,

unfortunately I did not keep the video (some GBs) for lack of space on my hard drive! The image I posted was processed from a stack of 200 frames - if you think this is still worth reprocessing I can email you the unprocessed tif-file.

Next time I get the chance to capture decent CaK I will definitely try out much larger stacks!

Thanks and regards,

Frank

Post Reply

Who is online

Users browsing this forum: ffellah and 4 guests