Discrete dynamics for convex and non-convex smoothing functionals in PDE based image restoration

In this article we consider a model that generalizes the Perona-Malik and the total variation models. We consider discretizations of this new model and show that the discretizations conserve certain properties of the continuous model, in particular convergence of the iterative scheme to a critical point and existence of a discrete Liapunov functional. Computational results are obtained that illustrate different features of the family of models.