彩色图像量化的FSCAMMD聚类算法

A Clustering Algorithm for Color Image Quantization (FSCAMMD)

提出了一种基于模式识别技术的彩色图像量化的新算法──基于最小距离最大 的快速统计聚类算法（FSCAMMD）。本算法克服了SCA算法对聚类中心初始值选取的不足，给 出了最大频度与类内最小距离最大相结合的方法──初始值优选法。实验结果表明，本算法 可较大幅度地减少图像量化后的总方差以及颜色失真度。

A new algorithm for color image quantization based on the pattern reco gnition technology is proposed in this paper. This is a fast statistical cluster ing algorithm based on maximizing minimum discrepancy (FSCAMMD). The algorithm c an overcome the shortcomings of the seeking method of initial value of the clust ering center of SCA algorithm, and gives a method of combining maximum frequency degree with maximizing minimum discrepancy, that is an optimum seeking method o f initial value of clustering center. The experimental results show that both th e total mean square deviation and lack fidelity of images quantized by the prese nt algorithm have a relatively big reduction.