Colour Image Noise Reduction Using Fuzzy Filtering

Abstract

Image filtering is one of the most important image processing tasks. Recently, fuzzy techniques offer a suitable framework for the development of efficient methods. In this paper, a recursive fuzzy filtering technique is presented to reduce additive Gaussian noise from color images using YIQ space. An adaptation scheme is developed to adapt the shape of membership functions of this filter according to the remaining amount of noise after each iteration. After the adaptation scheme, the fuzzy filter is based on two stages of fuzzy reasoning. The first stage computes a fuzzy derivative for eight different directions. The second stage uses these fuzzy derivatives to perform fuzzy smoothing by weighting the contributions of neighboring pixel values.
The simulation results show the effectiveness of this filter in reducing the additive
noise while preserving image edges very well. It is also favorably compared with the traditional
mean filter.