By Ctein
Last column, I told you how much fun Paula and I were having with our new Coronado PST solar telescope and some of the tricks I've come up with to squeeze even more enjoyment out of it. Well, within a few days, I figured out that it was really easy to make solar photographs through the PST! I really do mean easy, not just "easy for Ctein."
You need a digital camera that has raw mode and manual focus and exposure settings; your camera is not likely to be able to focus on the monochromatic red light and relatively featureless disc of the sun, nor is it likely to accurately meter. Use an eyepiece with a rubberized eyecup, common on shorter focal length eyepieces, so that you can press the lens up against the eyecup and keep out stray light (Fig 1).
Set the camera on manual and focus the lens at infinity. ISO 800 and 1/125th sec seem to work well with my 5mm (shown in figure 1) and 3.2 mm eyepieces. Your mileage may differ. Leave the aperture wide open. If your camera has a burst mode, turn it on. Focus the telescope so the image looks sharp to you. Center the features you want to photograph in the eyepiece. Press the camera lens against the eyepiece and hold down the shutter release to fire off a burst of photographs; the Olympus Pen will do 10–12 raw photos before it starts to slow down.
Make a whole lot of photographs! You'll see why.
Okay, that was the easy part. Now it's a little more work. Don't try to evaluate the photographs in the screen on the back of the camera; the deep red light falls well outside the gamut of the display, and they're going to look featureless, a lot like figure 2, top. This is also how they'll look with the default settings in ACR (figure 3, left). Settings more like on figure 3, right, will give you an image like the one at the bottom of figure 2. Eliminating most of the color cast and increasing contrast lets you see a lot more of the detail in the prominences and what's on the surface of the sun.
Experiment! This column is not a tutorial and I'm not explaining exactly what the settings mean nor which ones will work best for you. Play around for yourself! I use different settings for bringing out detail in the prominences and in the surface of the sun. I usually kick the contrast way up to see solar detail, and that pushes prominences into black (figure 4). Not shown in this screenshot is that I also made the curves a lot steeper to increase contrast. Nice thing about raw; you can process photos any way you want.
Figure 4 shows one of the November 17 photographs I referred to in my comments to the previous column. After massaging the photo like this, I could see much, much more detail in the sun than I could through the telescope. This is a much better "view" that I could get with the naked eye. Notice all the "floaters! Almost all those dark spots and clouds are bits of dust and dirt and smudges in the eyepiece. You need to clean your eyepiece scrupulously or your photographs will be filled with these. Even Thierry's photographs aren't spotless. But, the narrow U-shaped cloud near the limb of the sun, lower right, is real. I could tell because from frame to frame it moved with the sun; floaters stay in the same place in the field.
So, why make a lot of photographs? Atmospheric turbulence! From frame to frame, the heat ripples in the air will make big changes in the sharpness of your photographs. 1/125th sec is enough to freeze those ripples, but it won't make them go away. Some frames will be extremely blurry; only a few will be maximally sharp. You will need to look through dozens of frames to find the best one.
Figure 5 shows one of the best frames on November 18. This was a very nice small prominence, but the seeing was really awful. I could barely make out the thin "rind" of darker spicules along the edge of the sun. Normally that rind would be very clear, even though this scope is too small to resolve individual spicules. I picked that frame out of about 200 photographs I made, and it's still not very sharp. Yes, searching is very tedious. Professional solar astronomers have the same problem—that's what students are for.
While you can expect photographs to show you a lot more detail than you can see in plages, granulation, and sunspot groups, you'll see less in prominence photographs. Our brains are really good at picking out fleeting fine detail in a wavering image; much better than the camera.
Can we do anything about that? Yes, but now it's no longer easy. You'll need a good image sharpening program, such as FocusFixer and a user-controllable noise reduction program such as Noise Ninja. (Again, I'm not going to tell you how to use these—not enough time, not enough space. I repeat, experiment!)
I took figure 5 and applied FocusFixer to it with a 16 pixel radius. This gave me the best sharpness with the fewest artifacts, but the photo was covered with "orange peel" texture; an artifact of the extreme enhancement. I used Noise Ninja with a manual selection of areas in the sky and the surface of the sun to eliminate most of the orange peel. Figure 6 shows me as much detail as I could ever hope to get under the really awful seeing conditions. Notice how ragged the limb of the sun is and the "rind" is almost invisible. But it's almost (not quite) as much detail as I could see with the naked eye.
I'm waiting for a day with good seeing and a great prominence. It'll happen; every day is different. That's also part and parcel of real solar astronomy.
Maybe you already know this, but the best sharpness enhancing technique in astro photography is "stacking" multiple exposures together.
Free stacking software like Registax is available to produce a final picture that has greatly reduced noise and improved sharpness. It works considerably better than trying to sharpen single images using the usual photographic software.
Because you shot a burst of frames of the sun, you should already have a number of pictures that would work. Just edit out the obvious dogs, and feed the rest to Registax.
I assume stacking techniques would also work with any subject in which there is no relative motion (i.e. landscapes on a still day).
Posted by: William Schneider | Sunday, 29 November 2009 at 04:08 PM
Oh, now I have to get one of these things. Its your fault!
Have you tried stacking images in Registax or something similar?
Posted by: Bourquek | Sunday, 29 November 2009 at 04:15 PM
Sharpness issue? See here ! http://astrosurf.com/legault/atlantis_hst_transit.html
Mark
Posted by: Mark | Sunday, 29 November 2009 at 06:43 PM
Dear W & B,
Yes indeed, stacking is a promising technique for photographs such as these, but I haven't had the time to play with it yet.
Not only would it be valuable for improving sharpness, but I am fighting noise. As I said, the light one gets thru a scope such as this is not very bright, and I don't dare use very long exposures because of the atmospheric turbulence problem. So, even when I get a very sharp frame, low contrast detail (of which there is plenty on the surface of sun and in prominences) gets lost in the noise.
I will be giving it a try at some point. Probably will wait for a nice combination of good seeing and attractive solar features to inspire me.
~ pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com
======================================
Posted by: ctein | Sunday, 29 November 2009 at 07:49 PM
@William, I'm not an astronomer, but the surface features of the sun move. Can't stack images of subjects that are moving relative to one another and changing shape.
Posted by: Janne | Sunday, 29 November 2009 at 07:58 PM
Dear Mark,
Yes, indeed! These nicely illustrate variations in seeing. The top transit is not terribly sharp at all-- probably a limiting resolution in the same range as my dinky little scope (3 arc-sec) can achieve.
The middle transit nicely shows the pernicious nature of turbulence. The underlying granularity on the surface of the sun is pretty uniform, in reality. The smeary bands overlaying the enlarged photo illustrate how the seeing can very from point to point even in a single frame.
The bottom transit is wonderfully sharp right where the shuttle is-- definitely sub-arc-sec resolution. But if you pull up the full-size JPEG, you can see how lucky Thierry was; in that frame, the image of the shuttle fell on a tack-sharp zone right between two big smeary bands of turbulence.
Luck favors the prepared... but solar astronomy involves LOTS of luck.
pax / Ctein
Posted by: ctein | Sunday, 29 November 2009 at 09:12 PM
Janne...
See this page for some stacked images of solar prominences...
http://www.astronomie.be/registax/ericspage.html
The sun's features change slowly enough for the technique to be useful.
Posted by: William Schneider | Sunday, 29 November 2009 at 09:31 PM
"Focus the telescope so the image looks sharp to you"
If the camera lens is set to infinity but your vision is nearsighted, like mine, wouldn't the eyepiece be focused too far in for the camera lens to focus correctly?
Posted by: John Robison | Sunday, 29 November 2009 at 10:03 PM
Dear John,
I'm very nearsighted, and I do the 'scope focusing without my glasses. Works fine.
pax / Ctein
Posted by: ctein | Monday, 30 November 2009 at 12:58 AM
This kind of photography involves either an aesthetic subtlety or scientific peculiarity that I don't comprehend; or perhaps "grok" would be the better word. Since resolution seems to be key, it would appear that the overall graphic-ness of an image isn't important (I rather like the top image in fig. 2, as a graphic, that might go on (perhaps) a bottle of Japanese mineral water.)
And since the photographed features are much better done elsewhere, no particular scientific curiosity is being satisified. I conclude, then, that one gets from this the same engineering/quasi-scientific pleasure derived from shooting off amateur rockets to see if you can recover an intact egg from the nose cone...
Understand, I don't object to any of this, as it seems far less harmful than, say, an interest in methamphetamine or under-wired bras, but, I just don't understand. On the other hand, I have a good friend who breeds hostas, perhaps the most inane plant on earth. I don't understand that, either, yet he remains a good friend.
JC
Posted by: John Camp | Monday, 30 November 2009 at 02:05 AM
Perhaps something like http://research.microsoft.com/en-us/um/people/jiansun/papers/Deblurring_SIGGRAPH07.pdf>this would be an alternative to stacking?
(Admittedly the blur kernel due to turbulence is very different from that due to camera motion.)
Posted by: expiring_frog | Monday, 30 November 2009 at 03:39 AM
Dear John,
Yeah, think of it like model rocketry. Just something that some people find cool to do for onesself. It does not actually accomplish anything of note. Rather like making tourist snapshots of famous scenes while on vacation. People know they can buy better ones. Some people do, even.
I thought it was pretty cool how easy it turned out to be to make photos through this telescope. Then, of course, I got wrapped up in the geekery of wondering just how good I could make those photos come out. Saner people would stop at Step 1.
pax / Ctein
Posted by: ctein | Monday, 30 November 2009 at 01:34 PM
EP,
No, this isn't going to work.
The reason is that the method used in the SIGGRAPH paper requires a very sharp, albeit noisy from underexposure, image to provide the reference data for sharpening up the properly exposed photo. No such reference photo would exist for solar photos.
The problem is that there are two aspects to atmospheric blurring. One is, indeed, dynamic-- constantly moving air that makes the image shimmer and shake. Exposures longer than fraction of a second will get blurred by this kind of 'motion blur.'
Once you get down into the 1/60th-1/250th of a second range, you're freezing out that motion. What you get is kind of like photographing through a sheet of rippled glass. The photo is nice and sharp... but the image of what's behind the glass isn't.
pax / Ctein
Posted by: ctein | Monday, 30 November 2009 at 01:42 PM
Pictures of the sun scare me..so do kerosene heaters.
Posted by: David | Monday, 30 November 2009 at 02:58 PM
Cool
Posted by: Mike Plews | Monday, 30 November 2009 at 06:06 PM
@Ctein: Ah, I see, thanks for the clarification. I had assumed that the blur was entirely dynamic and hadn't considered that turbulence would give a distorted/blurred picture even at very short exposures.
Posted by: expiring_frog | Tuesday, 01 December 2009 at 12:53 PM
I purchased a 40mm Coronado PST in August 2008. Since then I have captured hundreds of photo's of prominences, flares and occasional sunspot groups.
I do not use a fancy digital camera, just a humble SPC900NC webcam which I purchased for £35 from my local Maplin store.
Video capture is easy using HandyAvi (shareware $30). A short 200 frame movie is made to capture the prominences and then another exposed for the disk detail. The individual frames of both movies are then stacked using Registax5 (freeware) and the two separate images are composited with Photoshop Elements 4.
Below are links to my PST albums on Facebook and Flickr -
Facebook:
http://www.facebook.com/album.php?aid=2015736&id=1242938734&l=a88243b836
Flickr:
http://www.flickr.com/photos/astronomer_dave
Looking forward to following your PST imaging and updates on here.
Kindest regards,
Dave
'@Astronomer_Dave' on twitter
Posted by: David Evans | Sunday, 06 December 2009 at 11:13 AM