Introduction

Since the discovery of wavelet transforms, sparsity became a principal tool in signal and image reconstruction. Gradually, the Fourier-based linear methods are being replaced by powerful nonlinear alternatives. We consider the problem of image estimation from incomplete or inaccurate measurements, utilizing sparsity to get improved error performance.

Main Contribution

We developed efficient signal-estimation algorithms based on the newly developed theory of sparse stochastic processes. The algorithms are based on statistical interpretation of the data to achieve error performance superior to current state-of-the-art algorithms.