On selection of centroids of fuzzy clusters for color classification

TL;DR

This paper introduces a new initialization method for fuzzy c-means clustering tailored for color classification, focusing on selecting vivid, distinguishable colors as initial centroids to improve clustering accuracy.

Contribution

It proposes a novel approach that extracts dominant colors and uses fuzzy membership to select initial centroids, enhancing fuzzy clustering for color classification.

Findings

Improved clustering accuracy for color classification.

Effective initialization method based on dominant colors.

Enhanced distinguishability of color clusters.

Abstract

A novel initialization method in the fuzzy c-means (FCM) algorithm is proposed for the color clustering problem. Given a set of color points, the proposed initialization extracts dominant colors that are the most vivid and distinguishable colors. Color points closest to the dominant colors are selected as initial centroids in the FCM. To obtain the dominant colors and their closest color points, we introduce reference colors and define a fuzzy membership model between a color point and a reference color.