An improved noise removal model based on nonlinear fourth-order partial differential equations

In this paper, we propose an improved noise removal model based on a nonlinear fourth-order partial differential equation (PDE). It associates with the minimization of a certain energy subject to spatially varying constraints involving local variance measures. We discuss the existence and uniqueness of the solutions for the proposed model. The main advantage of the proposed method over the related methods is that it can not only preserve textures but also avoid the staircase effect in smooth regions in the process of denoising. Experimental results illustrate advantages of our proposed method in visual improvement as well as an increase in the signal-to-noise ratio over related PDE methods.