The Improvement of ROF De-noising Model Based on AOS and Fourth-order PDE

Abstract

The paper is mainly to improve the de-noising model in Rudin-Osher-Fatemi (ROF) images. Firstly, discrete values of the nonlinear total variation ROF model based on Add Operation Splitting (AOS) method were proposed. Because the second order de-noising model has the obvious ladder effect in ROF de-noising model, this paper has combined fourth-order partial differential equation (PDE) and ROF model of AOS algorithm to optimize the de-nosing model. The experimental results show that the improved model has effectively achieved the double optimization of speed and the efficiency.