A Denoising Algorithm Based on PDE Variational Approach Using PM Potential

In this paper a new denoising algorithm based on PDE variational approach is proposed. The proposed algorithm uses the PM potential instead of the total variation norm in the ROF scheme. It utilizes both advantages of the traditional ROF and PM methods. The proposed algorithm reduces the isolated points of the PM and the staircasing effect of the ROF algorithms. To evaluate the proposed algorithm several experiments are performed which confirm the high performance of the proposed algorithm.