Noise filtering and detecting edge image in nonlinear anisotropic diffusion

Abstract

Nonlinear Anisotropic Diffusion (NAD) is a smoothing while edge preserving  process, which itself depends on local properties of the image. In order to avoid blurring of edges, automated problem for detecting edge pixels also differentiate between edge noise is important in NAD method. The main contribution in this paper to introduce an active stopping-function instead of Perona-Malik’s  stopping-function and Tukey’s stopping-function in NAD model. Numerical experimental results are demonstrated for comparing effective of these  stoppingfunctions to image quality evaluation criteria.