Image Processing Fundamentals

SNR

Thermal noise (Dark current)
Photon noise

As described in Section 6, in modern camera systems the noise is frequently limited by:

* amplifier noise in the case of color cameras;

* thermal noise which, itself, is limited by the chip temperature K and the exposure time T, and/or;

* photon noise which is limited by the photon production rate and the exposure time T.

Thermal noise (Dark current)

Using cooling techniques based upon Peltier cooling elements it is straightforward to achieve chip temperatures of 230 to 250 K. This leads to low thermal electron production rates. As a measure of the thermal noise, we can look at the number of seconds necessary to produce a sufficient number of thermal electrons to go from one brightness level to the next, an ADU, in the absence of photoelectrons. This last condition--the absence of photoelectrons--is the reason for the name dark current. Measured data for the five cameras described above are given in Table 10.

Table 10: Thermal noise characteristics

The video camera (C-5) has on-chip dark current suppression. (See Section 6.2.) Operating at room temperature this camera requires more than 20 seconds to produce one ADU change due to thermal noise. This means at the conventional video frame and integration rates of 25 to 30 images per second (see Table 3), the thermal noise is negligible.

Photon noise

From eq. we see that it should be possible to increase the SNR by increasing the integration time of our image and thus "capturing" more photons. The pixels in CCD cameras have, however, a finite well capacity. This finite capacity, C, means that the maximum SNR for a CCD camera per pixel is given by:

Capacity-limited photon noise -

Theoretical as well as measured data for the five cameras described above are given in Table 11.

Table 11: Photon noise characteristics

Note that for certain cameras, the measured SNR achieves the theoretical, maximum indicating that the SNR is, indeed, photon and well capacity limited. Further, the curves of SNR versus T (integration time) are consistent with equations and . (Data not shown.) It can also be seen that, as a consequence of CCD technology, the "depth" of a CCD pixel well is constant at about 0.7 ke^- / um².