Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (31): 181-183.DOI: 10.3778/j.issn.1002-8331.2010.31.050

• 图形、图像、模式识别 • Previous Articles     Next Articles

Self-adaptive color image quantization algorithm based on Fisher discrimination

HOU Yan-li   

  1. Department of Computer,Shangqiu Teachers College,Shangqiu,Henan 476000,China
  • Received:2009-04-20 Revised:2009-06-15 Online:2010-11-01 Published:2010-11-01
  • Contact: HOU Yan-li

基于Fisher判据的自适应彩色图像量化算法

侯艳丽   

  1. 商丘师范学院 计算机科学系,河南 商丘 476000
  • 通讯作者: 侯艳丽

Abstract: A self-adaptive color image quantization algorithm based on Fisher discrimination is studied.Firstly,the original image is quantized to 256 colors using the octree algorithm.Secondly,based on the quantitative relation of the NBS distance and the color difference of human visual,the initial clustering centers and number are determined automatically.Thirdly,a peer group corresponding to the initial clustering center is automatically achieved by using the Fisher discrimination.And then,a color image quantization effect is achieved.At last,simulations are performed on the presented algorithm,and the simulation result shows that the presented algorithm not only can solve the problem of giving the number of quantization in advance but also has better quantization effect than the octree algorithm and k-means algorithm with the same quantization number.

Key words: image, quantization, Fisher discrimination

摘要: 提出了一种基于Fisher判据的自适应彩色图像量化算法。首先用八叉树算法把原始图像量化为256种颜色,然后根据人类的视觉特性,参照NBS距离与人类视觉对颜色差别的定量关系,自动确定初始聚类中心及聚类数目,在此基础上,用Fisher判据自动确定出初始类中心的一个同组,从而实现图像的量化。实验结果表明所提算法无需事先给定颜色量化数目,在量化数目相同的情况下,量化效果明显优于八叉树算法和k均值算法。

关键词: 图像, 量化, Fisher判据

CLC Number: