Modeling and simulation of a spectro-polarimetric lenslet array imager

Abstract

A new, economical, lenslet-array-based imaging sensor design is proposed, simulated, and analyzed. In this investigation a bare lenslet array model is first developed in Code V®. The results show that, as expected, intolerable optical cross-talk is present in this simple system. This problem has been addressed in previous systems via the inclusion of a physical image separation layer. The alternative system proposed here to alleviate crosstalk involves the introduction of both polarizers and spectral filters. As a consequence this simple system design also provides spectro-polarimetric resolution. Simulations were developed in order to analyze the system performance of two designs. The simulation results were analyzed in terms of a measure of signal-to-noise ratio (SNR) and in terms of an en-squared energy that includes all subimages. The results show that a design employing only a few spectral filters suppresses crosstalk for objects of small angular extent but does not suppress crosstalk to a tolerable level for 2π steradian illumination, as evidenced by SNR less than one. However, the inclusion of more spectral filters results in a spectro-polarimetric thin imager design that suppresses crosstalk and provides finer spectral resolution without the inclusion of a signal separation layer.