Maximum Entropy Image Restoration Using Nonlinear Programming

A blurred and noisy image usually can be processed digitally to reduce its blur and noise level. Maximum entropy restoration has proved very useful in digital image processing. Frieden's solution by Newton-Rapson methods is limited by the small number of points in the one-dimensional case. In this paper, the formulation of maximum entropy approach in terms of an optimization problem is presented. The optimization problem is solved with linearly constrained nonlinear programming.