Friday, April 27, 2012

Picturing Mars

One of the perks of volunteering as a telescope operator at the Chabot Space and Science Center is, obviously, the telescopes. In particular, I work Saturday nights, and this particular Saturday night (21 April 2012) I was working with Rachael, the 20inch refractor. The moon was close to new, and Mars was high in the sky. It was about as good a night as any for viewing Mars, so once we kicked the last of our visitors out (around 11:00pm that night) I took over the telescope for myself. Woohoo!
The short story--after an hour of fiddling with adapters, experimenting with exposures, and collecting images; and another hour at home processing the collected data into an image, I came up with this result:
Mars through the 20" refractor at Chabot Space and Science Center.
Pretty nice, eh? You can see significant color variation on the surface, and even hints of a polar ice cap just to the right of the top edge.
 It was actually easier to get this image then one usually expects, and I attribute this ease to the rather awesome telescope aperture. The telescope is a 20inch refractor, with a prime focus at 28feet. The camera that I used is my own Nikon D80 camera body mounted at the prime focus. No other optics were used for this image. This image is actually constructed from a dozen individual images of the same view. Each image is a little darker than this, and a lot grainier, but otherwise looks much like the final product.

Unprocessed image, complete with ISO noise.
But the main problem with the unprocessed images is the graininess. Digital photographers would recognize the graininess as high-ISO noise, and a digital photographer would eliminate it by reducing the ISO setting and increasing either the aperture or the exposure time. Of course, the aperture is fixed. Can't change that. But I can certainly increase the exposure time. And I tried increasing the exposure time. The results were a much more blurry image. Why?

Well, if you look through a telescope using your eyes, you'll notice that the image is usually not perfectly still. You can see it jumping around a bit. That jumping around is caused by turbulence high in the Earth's atmosphere. If you take a long exposure, all that jumping around during an exposure appears as motion blur. By taking a fast exposure, blur from the jumping image is reduced, at the cost of a darker image. I tried a longer exposure, and there was too much motion blur. So that plan is thwarted. The image has ISO noise. If I reduce the ISO setting, it gets too dark, and neither aperture nor exposure changes can help.

So the technique I chose to use is called "stacking". I went ahead and used the high ISO setting and took a dozen images at fast exposures. The exposures were fast enough to reduce (although not eliminate) the motion blur but long enough to get a reasonably bright image. All the images were taken with the same setting. Then I used stacking software to align all the images and digitally stack them, like stacking negatives of exposed film. The software added the images together, then divided the image by the depth of the stack. The actual, stable image data adds up strongly, but the random noise only combines weakly. Thus, the exposed view is enhanced more then the noise, and it winds up looking like the first image.

This stacking technique has the effect of adding all the exposures of all the images together. But it is better then just taking a long exposure because I can cancel out the jumping of the image by shifting each image to cancel the jump in the image. Within the time of each exposure, the image moves very little, so there is little motion blur. Between each exposure the image may have jumped a lot, but I can correct for that by shifting the image. And there you go!

The main image is a stack of 26 exposures of 5mS per exposure. The stacked image was scaled up 2 times, and cropped to size.

No comments:

Post a Comment