Conjugate gradient acceleration of maximum-likelihoodimagerestoration

Abstract
A maximum-likelihood iterative deconvolution algorithm is accelerated using a conjugate gradient technique with a minimum mean square error criterion. Significant acceleration over previous techniques is achieved with little additional computation. Restorations with and without noise are presented which indicate the number of iterations required reduces from O(N) to O(√N).

This publication has 2 references indexed in Scilit: