3.2 CMOS detectors
The rival technology for CCDs is loosely called CMOS. That's not, at first sight, descriptive of the way the devices function, but it does identify the kind of production line that can be used to make them. You should recall that CMOS is the basis of mass memory. The commercial convenience of being able to produce imaging chips on a memory line has played a part in opening up the market for digital imaging, in cameras, mobile phones, remote monitoring systems, etc.
The image capture array is essentially the same as in CCDs, but instead of marching the data out through a single amplifier, CMOS amplifiers are built into each pixel, completing the conversion from charge to voltage locally. A CMOS amplifier is made from MOS transistors and so their inclusion is entirely compatible with the processing technology required for the light-capture components. However, a significant fraction of array space is given over to this local data handling, so again microlenses are incorporated to compensate for the loss of capture area. Figure 9 shows a diagram of a CMOS imager.
A disadvantage for the CMOS array is that each pixel uses a different charge-to-voltage amplifier, so there is additional statistical variation between pixels, as compared with CCDs. On the other hand, data can be drawn out of the array in parallel: a CMOS array can be interrogated row by row, as discussed in connection with CMOS memory. So, there is a trade-off between quality and speed.