计算机应用 ›› 2014, Vol. 34 ›› Issue (3): 828-832.DOI: 10.11772/j.issn.1001-9081.2014.03.0828

• 虚拟现实与数字媒体 • 上一篇    下一篇

融合KL散度和移地距离的高斯混合模型相似性度量方法

余艳1,2   

  1. 1. 华中科技大学 计算机科学与技术学院,武汉430074
    2. 武汉科技大学 理学院,武汉430065;
  • 收稿日期:2013-11-01 修回日期:2013-12-02 出版日期:2014-03-01 发布日期:2014-04-01
  • 通讯作者: 余艳
  • 作者简介:余艳(1980-),女,湖北襄樊人,讲师,博士研究生,CCF会员,主要研究方向:计算机视觉、机器学习。
  • 基金资助:

    冶金工业过程系统科学湖北省重点实验室(武汉科技大学)开放基金资助项目

Similarity measure method of Gaussian mixture model by integrating Kullback-Leibler divergence and earth mover's distance

YU Yan1,2   

  1. 1. School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan Hubei 430074, China
    2. 武汉科技大学 理学院,武汉430065;
  • Received:2013-11-01 Revised:2013-12-02 Online:2014-03-01 Published:2014-04-01
  • Contact: YU Yan

摘要:

为提高高斯混合模型(GMM)间相似性度量方法的计算效率和准确性,通过对称化KL散度(KLD)并结合移地距离(EMD)提出一种新的相似性度量方法。首先计算待比较的两个高斯混合模型内各高斯成分间的KL散度,对称化处理后用于构造地面距离矩阵;然后用线性规划方法求解两个高斯混合模型间的移地距离作为高斯混合模型间的相似性度量。实验结果表明,将该相似性度量方法应用于彩色图像检索,相对于传统方法能够提高检索的时间效率和准确性。

关键词: 图像检索, 高斯混合模型, KL散度, 移地距离, 颜色空间分布

Abstract:

To improve the computation efficiency and effectiveness of the similarity measure method between two Gaussian Mixture Models (GMM), a new measure method was proposed by means of integrating symmetrized Kullback-Leibler Divergence (KLD) and earth mover's distance. At first, the KL divergence between Gaussian components of the two GMMs to be compared was computed and symmetrized for constructing the earth distance matrix. Then, the earth mover's distance between the two GMMs was computed using linear programming and it was used for GMM similarity measure. The new measure method was tested in colorful image retrieval. The experimental results show that the proposed method is more effective and efficient than the traditional measure methods.

Key words: image retrieval, Gaussian mixture model, KL divergence, earth mover’s distance, color spatial distribution

中图分类号: