Title: A fuzzy rule based adaptive centre weighted median filter

Authors: Tung-Chien Chiang, Chih-Hui Chiu, Wen-June Wang

Addresses: Department of Electrical Engineering, National Central University, Jhongli, 320, Taiwan; Department of Electronic Engineering, Ching Yun University, Jhongli, 320, Taiwan. ' Department of Electrical Engineering, Chung Yuan Christian University, Jhongli, 320, Taiwan. ' Department of Electrical Engineering, National Chi Nan University, Puli, 545, Taiwan; Department of Electrical Engineering, National Central University, Jhongli, 320, Taiwan

Abstract: In this paper, a fuzzy rule based adaptive centre weighted median (FCWM) filter for the image restoration is introduced. According to some statistic measures of the input image, the parameters of fuzzy membership functions are determined to construct a set of fuzzy rules. The adequate centre weight is the output of fuzzy rules in order to attain the final output pixel, which is adaptive with the sliding of the local filtering window in raster scanning order across the image. Unlike the previous detection based filters looking for the threshold iteratively, the proposed filter establishes the associated rules quickly and in a straightforward manner. A lot of computer simulations show that the proposed FCWM filter can achieve a superior performance either in detail preservation or in noise suppression.

Keywords: identity filters; centre weighted median filter; CWM filter; fuzzy rule base; fuzzy logic; fuzzy control; image restoration; simulation; detail preservation; noise suppression.

DOI: 10.1504/IJCAT.2006.011145

International Journal of Computer Applications in Technology, 2006 Vol.27 No.2/3, pp.192 - 203

Published online: 19 Oct 2006 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article