A Nonlinear Hybrid Diffusion Model for Image Denoising

In this paper, a denoising technique is proposed based on a nonlinear anisotropic diffusion model with two diffusivities parameters, i.e., Charbonnier and total variation. These diffusivities are dependent on the diffusivity function for balancing. Furthermore, proof of the existence and uniqueness theorem of the model are presented. The convergent iterative scheme is proposed for the diffusion model. To discretize the diffusion model, the finite difference method with forward-backward diffusivities is used. The numerical results are given in terms of peak signal-to-noise ratio (PSNR) as a metric.