Computations in astro-imaging

Abstract

This work involves two-stage approaches to enhancing the quality of images taken through the atmosphere. First, a control problem arising in adaptive-optics is discussed. The problem involves optimal real-time control of very fast-acting deformable mirrors designed to compensate for atmospheric turbulence and other image degradation factors, such as wind-induced telescope vibration (windshake). The surface shapes of the mirrors must change rapidly to correct for time-varying optical distortions. The second stage of compensating for the effects of atmospheric turbulence generally occurs off-line, and consists of the post-processing step of image restoration. Here, the work involves large-scale computations, using either simultaneous image of a natural guide star or a large ensemble of images corresponding to different atmospheric realizations, to deconvolve the blurring effects of atmospheric turbulence.