Fuzzy based diffusion coefficient function in anisotropic diffusion for impulse noise removal

Abstract

This paper provides the use of rule based fuzzy scheme to define a new diffusion coefficient function in anisotropic diffusion for impulse noise removal with edge preservation. This is achieved by expressing the small, medium and large labels of second order pixel differences in fuzzy format. An aggregated output membership function of percentage of noisiness is then obtained by selecting an optimal linguistic value of second order pixel difference during inference process. The pixels have been classified as homogeneous, edge and noisy pixels based on the degrees of noisiness of the output membership functions. To achieve desired smoothing of the impulse noisy images with homogeneous background, the new diffusion coefficient function in anisotropic diffusion is redefined to vary it in accordance with the degrees of noisiness of the output membership functions. The experimental results have been compared with existing anisotropic diffusion methods as well as advanced median filtering method. It is observed through experimental results that the proposed method works satisfactorily for images having impulsive noise density upto 50%. ...$15.00. pixels in image. This noise is generally termed as random valued impulsive noise (RVIN) [1] which affects the dynamic pixel luminance range of the image. If we denote the dynamic pixel luminance range of an observed image f as [pmin, pmax] i.e. pmin ≤ fij ≤ pmax where (i, j) is the pixel position in f . Then, the RVIN model is defined by the gray level of f at (i, j) as follows [1]:

A Review on Performance Analysis of PDE based Anisotropic Diffusion Approaches for Image Enhancement

“…These all are very useful in various fields of mathematical science and engineering where parameter settings are required. However, there are very few techniques available in the literature [35] [36] [55] [80] [84] [85] [86] where these soft computing tools have been utilized in anisotropic diffusion filters. This motivates to explore the properties of the soft computing based techniques to develop more approximated and optimized anisotropic diffusion model for specific image applications.…”