Comet McNaught (2006/P1) in the Daylight Sky
January 10, 2007
 |
So I was having no luck at all finding Comet McNaught in the evening or morning skies from the foothills of western North Carolina. On the 10th, a cold front came through. Dry air. Deep blue sky. Figured if I could see Venus in that sky – easy in 14x70 binoculars – I had a pretty good chance of seeing McNaught a magnitude or two dimmer. Worked out the offsets from the Sun for both Venus and the comet. Used Venus for my focus target Then swept for the comet (very carefully) with the Sun dimmed by trees. This I did with a 35mm Panoptic in a 5-inch F6 Astro-Physics refractor (27x). No problem. Found the comet easily. But kept losing it when I'd try to change to a higher magnification eyepiece. Focus was critical to being able to see a comet against the blue sky. I never did see it through the 14x70 Fujinons. Finally got a good look at it with a 16mm Nagler. Much structure visible and much suggested. The sight begged for a photo to see if I could tease the tail out of the image. It was hinted at visually in the bright blue sky, so I was sure it would be more spectacular if I could just get it on a chip. Kept losing the comet every time I tried exchanging the eyepiece for a camera.
|
FInally, I focused on the Sun through the camera (unmodified Canon 20D) using a solar filter. Measured the extension of the draw tube with a caliper micrometer. Put the eyepiece back in the telescope, focused again. Offset to the approximate location of the comet. Took the filter off. Got a chair off the porch. Determined to sweep until the comet set or I found it. I found it. Locked the axes of the Giro alt-az mount. Put the camera on the telescope. Set the approximate focus with the micrometer. Aha! Now I could see the comet on the ground glass. Tweaked focus. Shot a dozen RAW frames 1/2000 second each. Here's one of them:
|
 |
Can you find it?
Upper right.
Above a dust
mote on the IR cut filter.
Got to clean that thing soon.
Here, let me mark
it for you:
Some dot, huh?
Well, let me tell you, it's a lot more than we astrophoto types usually get
to work with.
I dragged all twelve raw images into MaximDL.
Promptly
ran out of memory while trying to combine them all (I have 2 gigabytes of
RAM on this machine).
So I processed them in two groups of six.
First: I aligned all images on the comet.
Then I averaged the first group of six images. Then the second group of six.
Then I added the two averaged images together (there was a lot of headroom).
Here's what we had then:
Not bad! Note that the dust went away during this step, too.
The telescope wasn't tracking the comet,
so the comet moved a little in the frame between exposures. When I combined
them all with the
comet's position shifted to stack,
the dust
got averaged right out while the comet came up nicely.
I helped the process along with a gaussian blur in Photoshop, but most of the
dust elimination was
a natural result of the stacking process.
The only detail we really care about is the sky right
around the comet. The sunlit pines are
just so much terrestial noise, right?
Next I isolated the area immediately surrounding the
comet and had Photoshop CS3 apply histogram
equalization to
the whole image based on the pixel values there:
Neatness counts, so let's blur the bizarre edges of the pines:
That'll do. I can think of some more aggressive
processing to try, but
there's non-astro work that needs to be tended to, too.
Here's a crop of just the comet presented at nearly full
scale rather
than as a small part of a downsized frame:
 |
Thank you, Bill Curran, wherever you are. That image processing seminar back in 1982 or 1983 was one of the greatest classes in my academic career.
Come here from Team Sisyphus?
|