Noise in MTF Measurements.

The effect of shot noise on the MTF of a lens when measured by means of an edge scan is treated both analytically and experimentally. It is shown that the variance in the MTF, introduced by the shot noise, establishes an upper bound on the magnitude of any systematic error introduced into the MTF measurement by the noise. Both the variance and the systematic bias in the MTF measurement due to the noise can be reduced to arbitrarily small values by controlling the electrical bandwidth and the scanner velocity. Expressions are given for the maximum scanner velocity and the minimum elapsed time for the scan as a function of the acceptable variance.