Median-rational hybrid filters for image restoration

Abstract

A new class of nonlinear filters is introduced, called median-rational hybrid filters (MRHFs), based on rational functions (RFs). The output is the result of rational operation taking into account three sub-functions. It is shown that every sub-function will preserve details within its sub-windows. The proposed MRHF filters have the inherent property that on smooth areas they provide good noise attenuation whereas on changing areas the noise attenuation is traded for good response to change. It is shown that a consistent reduction in the objectively measured mean absolute error and mean square error is obtained.