Convergence analysis of blind image deconvolution via dispersion minimization

Abstract

A new non-linear adaptive filter called blind image deconvolution via dispersion minimization has recently been proposed for restoring noisy blurred images blindly. This is essentially a two-dimensional version of the constant modulus algorithm that is well known in the field of blind equalization. The two-dimensional extension has been shown capable of reconstructing noisy blurred images using partial a priori information about the true image and the point spread function in a variety of situations by means of simulations. This paper analyses the behaviour of the algorithm by investigating the static properties of the cost function and the dynamic convergence of the parameter estimates. The theoretical results are supported with computer simulations. Copyright (c) 2006 John Wiley & Sons, Ltd.