The ability to reliably and quickly determine the resolution of the scanning electron microscope (SEM) is highly desirable becauseonly then can the level of performance which is being achieved be compared to that which was anticipated or specified. Further, in activities such as automated line width measurements, or in telemicroscopy, the ability to quantify the imaging performance provides the remote operator with the necessary information to monitor, document, and optimize the instrument set-up. However, compared to the equivalent task on the transmission electron microscope (TEM) an analysis of SEM performance is hampered by poor signal to noise ratios, and the absence of ideal ‘white noise’ samples.

The two dimensional Fourier power spectrum of an image provides a readout of the frequency content of the information present in the micrograph. From an analysis of the power spectrum the highest spatial frequency transferred in the image can then be determined and hence the resolution of the micrograph can be specified.