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

Valle de fuego Sunspot

  • Please log in to reply
16 replies to this topic

#1 PhotonJohn

PhotonJohn

    Viking 1

  • -----
  • topic starter
  • Posts: 937
  • Joined: 11 Dec 2007
  • Loc: Viera Florida

Posted 16 January 2020 - 02:20 AM

This is a re-do of an image from 4/17/2019. My images are beginning to smooth out, but still think I'm over processing. I keep reprocessing but seem to be stuck. Photoshop is a bit confusing to me. Not sure I'm processing in the correct Photoshop order. Is there a correct order?  This was AR 2738 captured with a Lunt LS-60 double stack and ZWO ASI 174 mm camera. SharpCap, for capture, AutoStakkert 3.0, Registax 6 and PS CC for processing. Click image for full size.

 

09_56_19ar6ps1 (10)jpg.jpg


  • paulsky, rigel123, hopskipson and 6 others like this

#2 RickV

RickV

    Vanguard

  • -----
  • Posts: 2,332
  • Joined: 17 May 2013
  • Loc: Ottawa, ON, Canada

Posted 16 January 2020 - 05:26 AM

Looks pretty good to me John.

 

My processing order is:

SharpCap for capture, PIPP for quality selection, AutoStakkert 3.0, PhotoShop CS4 with plugins: Topas Denoise and AstraImage (for wavelet sharpening, Super Contrast and Lucy Richardson Deconvolution).  I'll often denoise after each sharpening routine).

 

Also, I found that AS!3 does a poor job of selecting the best frames and now do that exclusively in PIPP- which at least chooses 50% of the frames that I think are the best.

 

Best,

Rick


  • PhotonJohn and MalVeauX like this

#3 rigel123

rigel123

    Hubble

  • *****
  • Posts: 16,919
  • Joined: 29 Jun 2009
  • Loc: SW Ohio

Posted 16 January 2020 - 07:16 AM

Looks good John, maybe a little dark on the limb but that kind of adds to the depth of field so the image does not look flat.  As to a correct order I would be interested to hear other opinions as well. Typically for me once I bring an image into PS it is for cropping and any final adjustments to the histogram to bring out details to my satisfaction.  I typically lean away from too much sharpening as it can result in artifacts and in your image I’d say you have reached a nice balance.


  • PhotonJohn and MalVeauX like this

#4 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 6,358
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 16 January 2020 - 07:44 AM

Looks good to me!

 

Everyone has a style, there's no correct way to do it. Display for scientific data and/or simply pleasing pictures are two different things too and some people get bent over the pleasing pictures sometimes and will have different opinions on how an image should "look." So keep that in mind and do what you like with your intention. I like to have a reasonable balance between accurate representation and pleasing image at the same time, so I tend to post multiples of the same image in different ways for the different purposes. Personally I like the softer processed images than the harder, over-sharpened ones, just personal preference. Composition matters too so I take that into consideration big time.

 

My work flow is pretty simple these days:

 

1) AS!3, Stack & align (I output 4 images from this process each time so I can evaluate the best of the 4 depending on the quality of the frames). I typically use noise robust of 6 with LaPlace on, cropped and analyze my entire video file (I do this in batches, where you select several videos at a time so it will do them all automatically). For output, I choose an alignment point based on image scale. This is often going to be 32~48 for me in AS!3. I generally output 61, 101, 181 and 251 frame stacks, I do not use percentage typically. I evaluate them. The graph in AS!3 is useful, but is not accurate, so I just use the actual output images to determine if the data was good or not. Really good data, I may go back and try different stack sizes (if it increases quality, or not). There's a lot of different ways to approach this, massive stacks with strong processing is one method, good data with smaller stacks and softer processing is another.

 

2) IMPPG to adjust levels and if the limb is involved I do a slight shadow lift that is targeted on the shades that include prominences (I do disc & proms with one shot these days). I will do a little deconvolution and unsharp mask, but I do it gently. If it looks "display level sharp" here, it's too sharp. So I back off.

 

3) Photoshop (CS5.1 for me, old). I crop and clean up gradients and/or any noise. I lift major proms. I suppress glow around the disc and proms. I do one or two high pass filters with very low opacity (this is my final sharpening routine). I will attempt a low opacity unsharp mask to see if the data can take it without over-sharpening. I keep these opacities in the 25~50% ranges, tops. Contrast & levels at the end (because if you do these first, sharpening routines are harsher as you clip into white data). Then I do any coloring, inverting, etc for fun.

 

Very best,


  • PhotonJohn likes this

#5 RickV

RickV

    Vanguard

  • -----
  • Posts: 2,332
  • Joined: 17 May 2013
  • Loc: Ottawa, ON, Canada

Posted 16 January 2020 - 11:17 AM

John, as Marty points out... a lot depends upon your seeing.

Marty's seeing is typically much better than mine.

I too can use IMPPG for sharpening but only rarely as my images are usually formed from a stack of just 20 best frames; IMPPG fails miserably with such a low stack count.  If I had a an image composed from 100+ stacked images then it works fine.  I found that AstraImage works better for me.

 

Marty... how do you, "suppress glow around the disc and proms"?

 

Best,

Rick


  • PhotonJohn and MalVeauX like this

#6 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 6,358
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 16 January 2020 - 11:26 AM

 

Marty... how do you, "suppress glow around the disc and proms"?

 

2~3% Shadow Burn with a large diffuse brush.

 

I often will use 2~3% Midtones Dodge with a small diffuse brush first.

 

Very best,


  • RickV likes this

#7 bigdob24

bigdob24

    Viking 1

  • *****
  • Posts: 636
  • Joined: 19 Apr 2008
  • Loc: Central Illinois

Posted 16 January 2020 - 11:37 AM

As a visual observer I know nothing about the process of getting one of the fantastic photos you guys share.

On average how long does it take to get the data you need and then how long does it take to process?

The only camera I’ve messed with was a Mallincam on my 24” Dob, way different but rewarding when it all came together and frustrating the rest of the time .

BD


  • PhotonJohn and MalVeauX like this

#8 PhotonJohn

PhotonJohn

    Viking 1

  • -----
  • topic starter
  • Posts: 937
  • Joined: 11 Dec 2007
  • Loc: Viera Florida

Posted 16 January 2020 - 11:45 AM

Thanks Marty, Rick and Warren. There is a boatload of info in your responses. Learning your scope, camera, capture and processing programs is overwhelming sometimes when your old brain is techno challenged. You three "Wise" men always post great images and give me insight into the solar game.  


  • rigel123 and MalVeauX like this

#9 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 6,358
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 16 January 2020 - 11:46 AM

As a visual observer I know nothing about the process of getting one of the fantastic photos you guys share.

On average how long does it take to get the data you need and then how long does it take to process?

The only camera I’ve messed with was a Mallincam on my 24” Dob, way different but rewarding when it all came together and frustrating the rest of the time .

BD

Depends on the image scale.

 

With my 0.5"/pixel to 1.5"/pixel image scales I can get the data in under 5 minutes any time of day as long as there are no clouds and process it in under 30 minutes ready to share. My 0.3"/pixel data can take longer to get waiting for good seeing moments, and generally takes longer to process because I do rapid bursts and the batch stacking and aligning takes longer. The processing after stacking and aligning takes me very little time typically, I can have it done in less than 10 minutes per image easily. What takes me longer is when I'm doing that on 8~12 images, so then it may be a few hours before I have everything done. I have my workflow down to a very rapid routine though and I do it often, so its second nature at this point.

 

My setup is always ready to go in my observatory, so I can walk out and begin imaging in 2~3 minutes typically. This is one of the biggest time savers.

 

Very best,


Edited by MalVeauX, 16 January 2020 - 12:04 PM.

  • PhotonJohn likes this

#10 PhotonJohn

PhotonJohn

    Viking 1

  • -----
  • topic starter
  • Posts: 937
  • Joined: 11 Dec 2007
  • Loc: Viera Florida

Posted 16 January 2020 - 12:00 PM

As a visual observer I know nothing about the process of getting one of the fantastic photos you guys share.

On average how long does it take to get the data you need and then how long does it take to process?

The only camera I’ve messed with was a Mallincam on my 24” Dob, way different but rewarding when it all came together and frustrating the rest of the time .

BD

 

It only takes less then a minute to capture a video on average. Because I'm new at solar it takes me about three or four hours to play with my images to get something decent to post. Marty takes about ten minutes (it seems like) to crank out a mess of them. My HOA will not allow a dome so I have to set up whenever GONG network shows something worth imaging. Never find it frustrating, always love to be out in the sun.


  • MalVeauX likes this

#11 bigdob24

bigdob24

    Viking 1

  • *****
  • Posts: 636
  • Joined: 19 Apr 2008
  • Loc: Central Illinois

Posted 16 January 2020 - 12:36 PM

Thanks for sharing.

For visual , my mount is in the garage so setting the scope on with Binos and rolling it out about 10 min.

BD


  • PhotonJohn and MalVeauX like this

#12 PhotonJohn

PhotonJohn

    Viking 1

  • -----
  • topic starter
  • Posts: 937
  • Joined: 11 Dec 2007
  • Loc: Viera Florida

Posted 16 January 2020 - 12:38 PM

Capturing in the middle of the day usually gives me problems with Newton rings even when using a tilt adapter. Later in the day after 2pm EST the rings usually disappear. I also have trouble capturing a decent flat in SharpCap. Any ideas on these issues?


  • MalVeauX likes this

#13 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 6,358
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 16 January 2020 - 01:04 PM

Capturing in the middle of the day usually gives me problems with Newton rings even when using a tilt adapter. Later in the day after 2pm EST the rings usually disappear. I also have trouble capturing a decent flat in SharpCap. Any ideas on these issues?

The rings are due to there being flexure or slop somewhere so that the beam of light is not square to the sensor. The tilt adapter usually can handle this, but if the tilt required is beyond the tilter's range, it will not. The reason you see more profound rings at various times of day likely has to do with the angle the scope sits and the distribution of weight and the moment arm generated reveals where that flexure or slop is happening (so mid day when your scope is pointing higher than late day for example).

 

SharpCap cannot give you a real time flat but it depends on how you want to accomplish it. The defocus method will work here, if you can fill the FOV with the sun based on  your image scale. I would fill the histogram to about 67% and do about 100 frames and stack them. Then don't change orientation of anything in your imaging train and continue to image with the similar histogram fill range (+/- a bit). Unfortunately SharpCap is more limited for solar due to the inability to do a real time flat so that you can see what it is. SharpCap is great as a tool for polar alignment, real time stacking and for time lapse imaging. Maybe real time flat calibration will come in later in a future release we hope? So until then you can make flats either way, but you'll have to wait until you process to see if they worked or not. 

 

I find FireCapture to be superior for solar imaging with its options, plugins and the real time ability to apply a flat frame on the video feed so that you can see real time and expose real time with the flat already there. This way you can do a defocused flat frame or you could do a diffuser based flat frame (using an opaque diffuse plastic material in front of the aperture into the scope to scatter light and make a flat frame that way, this works even with full discs showing void of space in the FOV, the key is placement of the material and the right material so that its diffusing and scatting light with no detail to be seen and it typically has lower transmission). Same thing applies here, defocus or diffuser method, exposure histogram fill to about 65%~67% (no gamma) and create the flat and apply it real time. Continue to expose the histogram to a similar level (+/- a bit) and again you cannot change orientation of any components after the flat is made.

 

Very best,


  • PhotonJohn likes this

#14 PhotonJohn

PhotonJohn

    Viking 1

  • -----
  • topic starter
  • Posts: 937
  • Joined: 11 Dec 2007
  • Loc: Viera Florida

Posted 16 January 2020 - 03:12 PM

Thanks Marty. Looks like it's time to learn FireCapture.


  • rigel123 and MalVeauX like this

#15 rigel123

rigel123

    Hubble

  • *****
  • Posts: 16,919
  • Joined: 29 Jun 2009
  • Loc: SW Ohio

Posted 16 January 2020 - 04:09 PM

What surprises me with Newton Rings is even with the auto flat being applied when I go to process is when they start to show up. That's really frustrating.



#16 RickV

RickV

    Vanguard

  • -----
  • Posts: 2,332
  • Joined: 17 May 2013
  • Loc: Ottawa, ON, Canada

Posted 16 January 2020 - 04:44 PM

On average how long does it take to get the data you need and then how long does it take to process?

BD

It takes me a while to gather data; seems I always have clouds!

 

Get the Data:

On a great day, I'll spend maybe 1/2 hour gathering multiple 30 second full-disc videos with my small scope and then maybe one hour and a half with the big scope going round and round the solar perimeter gathering multiple 10 second videos of proms.

 

Process the Data:

For a good day, like above, it takes me about two to four hours to process those videos down to the final 6 to 8 still images for posting.

 

It just depends... one image could take 10 minutes while another could demand a 1/2 hour!

 

Best,

Rick



#17 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 6,358
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 16 January 2020 - 04:56 PM

Heya,

 

Well, this finally prodded me to do a follow up tutorial on processing. I got some data this late afternoon to use, so why not!

 

If you have a minute, here's an 18 minute start to finish workflow on some data I acquired today this afternoon. It's fast paced, so if you're new to the lingo of the software and solar stuff in general, you may lose a few bits. But if you're already familiar with most of it, it's a fast way to quickly see my processing. This is from stacking to processing, start to finish. I took my time a bit with it to explain things and also included the entire boring time it takes to stack and align just to be completely transparent with time frames. Actual processing time is very short. But it shows how I use the software. I did this with a small aperture today, 100mm, so the results are more accessible to most people with respect to their expectations of what could be replicated to test things out. Forgive any mistakes or misspoken blurbs!

 

waytogo.gif

 

https://www.cloudyni...-jan-16th-2020/

 

Very best,
Marty


Edited by MalVeauX, 16 January 2020 - 05:27 PM.



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