A CCD, or charged coupled device, is basically a mosaic-layer of light-sensitive sensors known as photosites, or, more familiarly, pixels. These pixels are arranged into a flat, usually rectangular surface and an image from either a camera or a telescope lens is projected onto it. Each pixel gathers a tiny electrical charge, which varies in size according to how much light falls on it. This makes them ideal for astrophotography.
Once an image is captured, the charges are measured and converted into information by the camera’s circuitry. This information is sent to a processor, either at the time or later, where it’s converted into new pixels on screen, allowing us to see the image captured by the CCD.
These sets of information are stored in the computer as image files and there are many programmes that can process the files, allowing people to enhance or change the image in many ways.
Signal and noise
A CCD camera is much better at capturing very faint images than old film-using cameras. There are problems, however, mostly due to so-called “noise”.
We see the image because the pixels vary in brightness according to the light emanating from the object we want to photograph. Unfortunately, not all variations are due to the image we want. Variations that we want are known as signal. Variations caused by unwanted energy or causes are called noise.
Good images have a good signal-to-noise ratio; a poor signal-to-noise ratio leads to poor contrast, speckles or other flaws in the image.
Where does noise come from?
Noise will get into a CCD via dark current, heat, quantum noise, bias and other irregularities. Supercooled cameras specially developed for astrophotography can combat these problems, but not entirely. They still perform much better than regular digital cameras, which are very noisy indeed.
Bias, or offset
Each pixel will always have a minimum electrical charge, even if there’s no exposure whatsoever, so no pixel ever registers zero. This is bias.
In addition to bias, each pixel will also gather charge during the exposure, even if it doesn’t receive any photons, so each pixel value will be a bit higher than it should be and this dark current can be worked out by the length of the exposure.
Bias and dark current will make very pixel a little brighter than it should be, so calibration by dark frame subtraction is the answer.
Flat field inhomogeneities
Pixels vary in sensitivity across the CCD – some don’t respond and some are always registering at maximum. In addition, bias and dark current can vary; there’s also light and heat intrusion or leakage within the camera itself that can lead to bright spots. Then, there’s problems that are more familiar, like internal reflections and strange blobs caused by dust particles on the CCD window. Calibration comes in again to try to smooth out these irregularities.
Each pixel’s charge can vary simply because of the many physical processes taking place inside them. A pixel might not make the same value each time it’s exposed to the same amount of light. These variations also affect bias, dark current and signal; altogether, this makes for a grainy image with fine detail loss.
These are high-energy particles from space. They zoom in and hit anything in their way; when that something is a CCD, it results in a bright spot than can look like a star. If the spot isn’t there on other exposures of the same area of the sky, it’s probably a cosmic ray and you’re best to discard the image.