Department of Computer Science and Engineering,Sharif University of Technology
Abstract. Morphological ltering is a useful technique for the processing and analysis of binary and
gray scale images. The extension of morphological techniques to color images is not a straightforward
task because this extension stems from the multivariate ordering problem. Since multivariate ordering is
ambiguous, existing approaches have used known vector ordering schemes for the color ordering purpose.
In the last decade, many dierent color morphological operators have been introduced in the literature.
Some of them have focused on noise suppression purposes. However, none has shown good performance,
especially on edgy regions. In this paper, new color morphological operators, based on a fuzzy principle
component analysis, are proposed for noise removal. These operators employ statistical information
(obtained by applying a fuzzy clustering algorithm on the color space) to achieve the desired results for the
denoising application. The performance of the proposed operators is compared with recent morphological
operators, reported in the literature, for denoising purposes and the superiority of the proposed method is