The resolving power of an optical imaging system is limited by residual aberrations and diffraction effects. The Rayleigh-Abbe diffraction limit of resolution corresponds to radius of the central lobe of the point spread function of an aberration free diffraction limited system. An attempt to circumvent this limitation was proposed by Toraldo di Francia, who showed that suitable pupil plane filtering can overcome this resolution limit, albeit over a restricted field. This paper reports results of our investigations on the use of evolutionary programming to obtain globally or quasi-globally optimum solutions in synthesis of lossless Toraldo filters consisting of concentric unequal area zones of fixed phase.

Resolution capability of any optical imaging system is limited by residual aberrations as well as diffraction effects. Overcoming this fundamental limit is called super-resolution. Several new paradigms for super-resolution in optical systems use 'a posteriori' digital image processing. In these ventures the three-dimensional point spread function (PSF) of the lens plays a key role in image acquisition. A straightforward tailoring of the PSF can be performed by appropriate pupil plane filtering. With a brief review of the state-of-art in this research area, this paper dwells upon the inverse problem of global optimization of the pupil function by phase filtering in accordance with the desired PSF.

No abstract