Image clustering using higher-order statistics

被引：4

作者：

Rajagopalan, AN ^{[1
]}

机构：

[1] Indian Inst Technol, Dept Elect Engn, Madras 600036, Tamil Nadu, India

来源：

ELECTRONICS LETTERS | 2002年 / 38卷 / 03期

关键词：

D O I：

10.1049/el:20020092

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Traditional algorithms for clustering image data have used Euclidean or Mahalanobis distance, Here, a more general higher-order statistics-based closeness measure derived from a series expansion for a multivariate probability density function in terms of the Gaussian function and the Hermite polynomials is proposed for clustering. The superiority of this measure is demonstrated with an example application.

引用

页码：122 / 124

页数：3