一种基于核的聚类色彩量化算法

作者：王自立，何通能 日期：2008-02-25/span> 浏览：3426 查看PDF文档

一种基于核的聚类色彩量化算法

王自立，何通能
（浙江工业大学 信息工程学院，浙江 杭州 310014）

摘要：针对K均值聚类在色彩量化中的应用，提出了一种基于核的聚类色彩量化算法。将原空间中待聚类的样本通过非线性映射，映射到一个高维的核空间中，从而将非线性问题变为线性问题，并通过Delphi编程加以实现。实验结果证明，该算法计算简单，鲁棒性强，具有一定的实用价值。
关键词：色彩量化；K均值；核方法聚类；Delphi
中图分类号：TP301文献标识码：A文章编号：1001-4551(2008)01-0025-03

A color image quantization algorithm based on kernel clustering
WANG Zili, HE Tongneng
(College of Information Engineering, Zhejiang University of Technology, Hangzhou 310014, China)
Abstract: A color image quantization algorithm based on kernel clustering was proposed. The data was mapped from their original space to a high dimensional space so that nolinear problem may become linear one in feature space. The algorithm was realized with Delphi. The experimental results show that the algorithm has rather good performance with low complexity, robustness and highly practical value.
Key words: color quantization; Kmeans; kernel clustering; Delphi
参考文献(Reference):
［1］ZHOU B, SHEN J Y, PENG Q K. An algorithm based on color cluster feature for color ouantization［J］. MINIMICRO SYSTEMS,2004,25(11):1998-2001.
［2］JIANG X Y, BUNKE H. Edge detection in range images based on scan line approximation［J］. Computer Vision and Image Understanding,1999,73(2):183-199.
［3］周赳.纹织CAD系统的应用功能和研发方向［J］.纺织导报,2003（2）:77-78，80-101.
［4］赵燕伟,王万良.基于聚类分析的色彩量化新算法及其应用［J］.计算机辅助设计与图形学学报,2000,12(5):340-343.
［5］鲁力,颜钢锋.基于色彩量化的自动配色新算法及其应用［J］.计算机工程与应用,2003,39(20):110-112.
［6］吕建平,黄英.几种图像色彩量化方法的分析比较［J］.计算机应用,2002,22(5):86-88.
［7］龚斌,孟祥旭.一种新型图像颜色量化算法及其实现［J］.山东大学学报,2003,38(1):58-60.
［8］边肇祺,张学工.模式识别［M］.北京:清华大学出版社,2000.
［9］KORPARANOVA L G, RISCH T. Completing CAD Data Queries for Visualization［C］. International Database Engineering and Applications Symposium,2002:130-139.
［10］KRAUT S K, SCHARF L L, MCWHORTER L T. Adaptive subspace detectors［J］. IEEE Trans, on Signal Processing,2001,49(1):1535-1541.