The raw image comming from my camera is not a pretty sight, and as it is is worthless for scientific or asthetic use. Let me show you one such image shown at right.
You might think the image passable, but as you will see, it contains a large number of artifacts that have nothing whatsoever to do with the image.
The first artifact we need to remove is that of the camera structure itself.
This structure exists because of the electronic nature of the cce chip itself with its channels, readout mechanisms, and other structural elements.
These structurs generate heat, and because of this induce structural dark current onto the image. Fortunately we can take a picture of this bias.
I would consider neither of these images completely representative of what can be seen on the actual screen, but the bias frame is shown at right. You will notice the dark at the top of the frame gradually giving way to lighter tones toward the bottom. You will also see a couple pixels that are no longer working properly on this frame. This is not really part of the bias, it's just a defect on the chip.
The darkness of the image is deceptive because the image has been stretched so that it can be seen. The effect from top to bottom is small, the pixel ranges are from zero to 102 adu, with a standard deviation of about 3.4. Note that this is out of a maximum well depth of 65000, so the effect is quite small. It nonetheless should be subtracted from the image. It's effects cannot really be seen when this is done, but it does affect measurements later. No refined raw image is shown at this time since the effect is so small.
The bias frame is created by taking the shortest possible image with the shutter closed. The image is taken at the temperature the images will be taken.
The frame at right is called the dark frame. It is an image taken at the same temperature as the raw image, and one that exposed for at least as long, but with the shutter closed.
Notice how much the dark looks like the raw. It's just that the dark shows no image detail
The idea here is to get an image of the dark current produced by the camera at that temperature in the time the exposure is made. This dark current is then subtracted from the raw image to produce a refined raw image. Sometimes a number of dark images are stacked and averaged to produce a refined dark before subtraction. The one at right was constructed from three 310 second dark images.
The refined raw is shown at right. You can see a good bit more detail on the image now. Don't be concerned with the hash marks shown on the image. They are jpeg compression artifacts and don't appear on the originals.
Note the dark circles. These are dust donuts that have shown up after the gross dark current effects were removed.
The dust donuts are part of the optical path, and will be removed in a moment.
The last effects to be removed are the optical effects. These include optical issues such as dust, field brightness issues, and so forth.
To remove these artifacts we need to take a picture of the optical system, and then remove the optics from the image.
The image at right is such a picture. Notice the brightness differences between the corners of the image and the center. Notice the out-of-focus dust?
The flat is a short exposure image taken at twilight. I personally take a number of flats and then average them. The flat is taken to near saturation. I like my flats to average between 45000 and 60000 ADU.
I'll take as many as 9 flats. I'll also take a series f none or so short darks and as many biases.
Then I'll average the biases and subtract from each of the flats, then I'll average the darks and subtract that.
Then I'll normalize and average the flats to produce themaster flat as shown at right. Again don't mind the crosshair. It's not part of the image of the flat.
The image shown at right is the result of the removal of the flat. The flat is not subtracted as was done with the bias and the dark. It is divided into the refined raw instead.
The result is a nice flat field without dust donuts or center brightening.
Notice the detail of the image has improved considerably.
The image is a much more accurate description of the sky and has much improved scientific usefulness.
One last thing I will usually do to bring out the detail is to stretch the image and reverse to a white background, This improves the eye's ability to pick up detail.
It is from this reversed image that photometery is done.
