Convection diffusion model for image processing (Q2844637)

There are many mathematical models for image processing that are applied in computer visualization. More precisely, the nonlinear diffusion PDE has been applied in image processing. In the paper under consideration the authors propose a convection-diffusion filter by adding a convection term in the modified diffusion equation as a physical interpretation for removing the noise. The dynamics of these equations is investigated by the discretization of the convection-diffusion model via cellular neural networks (CNN). The numerical experiments here proposed show that the developed approach is reasonably better in removing noise in comparison with other methods.