Unlike their radio counterparts, optical astronomers have paid relatively little attention to the role of the aperture function when designing their high resolution imaging experiments. This is despite the fact that it can be easily modified with the use of a suitable pupil plane mask. While the defining characteristics of the two most favoured pupil configurations have never been in doubt – negligible atmospheric noise in the case of a non-redundant configuration of small holes and high photon rates for a filled pupil – the relative merits of these two choices in terms of imaging performance in the presence of turbulence appear not to have been carefully investigated until very recently.

Most existing comparisons of fully-filled aperture (FFA) and non-redundant mask (NRM) based imaging strategies appear to have ignored a number of fundamental and practical difficulties that are often encountered in practice. In the intermediate regime, between very high and very low light levels, that characterizes most astrophysical applications hybrid imaging schemes seem most profitable. These utilize partially-redundant pupil geometries that combine the advantages of redundancy at low light levels without incurring the penalties associated with fully redundant beam recombination. Such pupil geometries are also useful in reducing the level of systematic effects that often plague speckle imaging experiments.