SPACE-VARIANT RECURSIVE RESTORATION OF NOISY IMAGES

A new approach to the modelling and filtering problems of nonhomogeneo us images is proposed. The main assumption is that the whole image can be modelled by a collection of open subregions where the 2-D signal i s regular enough to be well described by a smooth two-dimensional Gaus sian process. The boundary points between two adjacent subregions are the image edges which represent sharp discontinuities in the signal di stribution. A non-stationary state-space representation with structura l information about image edges is then obtained. This allows the mode l to vary, adapting to the presence of spatial discontinuities. The re ported simulation results show high filter performances.