Complex-Valued Image Recovery from Multiple Measurements

Capable of all weather day-or-night imaging, spotlight mode airborne synthetic aperture radar (SAR) is widely used for surveillance and mapping. Obtaining practically artifact- and noise-free SAR images on which practitioners can rely is imperative, and while myriad computational methods have been developed to recover SAR imagery, several challenges remain. This study develops a new recovery algorithm based in rigorous mathematical analysis to address some of these challenges. Specifically, it ensures that phase information is not lost in the recovery process, which may be important in coherent downstream processing tasks. Further, an empirical Bayesian approach is adopted to improve the robustness of the recovery as well as to provide uncertainty quantification.