Blind image restoration by anisotropic regularization

This paper presents anisotropic regularization techniques to exploit the pi ecewise smoothness of the image and the point spread function (PSF) in orde r to mitigate the severe lack of information encountered in blind restorati on of shift-invariantly and shift-variantly blurred images. The new techniq ues, which are derived from anisotropic diffusion, adapt both the degree an d direction of regularization to the spatial activities and orientations of the image and the PSF. This matches the piecewise-smoothness of the image and the PSF which may be characterized by sharp transitions in magnitude an d by the anisotropic nature of these transitions, For shift-variantly blurr ed images whose underlying PSF's may differ from one pixel to another, we p arameterize the PSP and then apply the anisotropic regularization technique s. This is demonstrated for linear motion blur and out-of-focus blur. Alter nating minimization is used to reduce the computational load and algorithmi c complexity.