Image Processing Reference
some subject matter. If you are eager to get the best out of your imagery, spending some
time with Equilibrium DeBabelizer software and processing your pictures with different
settings will be worthwhile.
Digital Imaging Processes in the Camera
The imaging process in a digital camera is fundamentally different from that in a film cam-
era. The light is focused via the optics in the usual way, but the shutter is not a mechani-
cal device at all. It is an electronic impulse that samples a CCD imaging array. From this a
voltage is measured for each pixel and that is passed through an analog-to-digital con-
verter to generate a digital sample value. Because the behavior of the CCD is not respon-
sive in the same way as the human eye, a mapping function scales the values to enhance
the detail in the image. Figure 6-14 shows this general arrangement.
Camera resolutions are increasing and costs are reducing. It is now possible to get
cameras with 5 million pixels for just a few hundred dollars or less. Increasing the resolu-
tion does not materially help us with video cameras other than perhaps to offer high-def-
inition video cameras to consumers. High-definition CCD devices operated at SDTV
resolutions could use several imaging cells to sample a single pixel in order to average out
any noise artifacts.
Color Imaging with a Single-Chip Camera
Getting a single imaging chip to detect color requires the placement of a filter over each
picture element so that it only responds to the desired color. This is necessary because the
imaging sensor is monochromatic and cannot tell the difference between a red photon and
a blue one.
There are three techniques:
Multi pass, one exposure for each color. This is not practical for moving images.
Color filter arrays over a single sensor.
Beam splitting via prisms to multiple sensors.
Figure 6-14 How digital cameras work.