计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (35): 169-171.

• 工程与应用 • 上一篇    下一篇

基于改进Hausdorff距离的人脸匹配方法

刘福新1,3,杜世培1,陈益强2   

  1. 1.贵州大学 电子科学与信息学院,贵阳 550003
    2.中国科学院 计算技术研究所,北京 100080
    3.上海中科计算技术研究所,上海 201203
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-11 发布日期:2007-12-11
  • 通讯作者: 刘福新

Application of modified Hausdorff distance in face match

LIU Fu-xin1,3,DU Shi-pei1,CHEN Yi-qiang2   

  1. 1.College of Information Engineering,Guizhou University,Guiyang 550003,China
    2.Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100080,China
    3.Shanghai Institute of Computing Technology,Chinese Academy of Sciences,Shanghai 201203,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-11 Published:2007-12-11
  • Contact: LIU Fu-xin

摘要: 提出了一种基于改进Hausdorff距离的人脸相似度匹配的方法,该方法首先将人脸划分为脸型、双眼、鼻、嘴等几个特征点集,分别计算各部分的改进Hausdorff距离,然后进行加权计算相似度。利用该方法,在ASM(主动形状模型)定位人脸的基础上进行了人脸检索。实验表明,利用人脸相似度计算方法对人脸特征库进行搜索,达到了较好的效果。同时结合ASM自动人脸检测,本方法可以全自动完成人脸匹配,应用于人脸识别及数字娱乐等领域。

关键词: Hausdorff距离, 图形匹配, 人脸检索, 主动形状模型

Abstract: This paper presents a new method to match human faces with the Modified Hausdorff Distance(MHD).First we divide human face into four parts:face profile,eyes,nose and mouth,the MHD of each part is calculated.And each part is given a certain weighting value.Then with the advantage of the Active Shape Model(ASM),we extract the facial feature points and define a new similarity calculation between two faces based on the Modified Hausdorff Distance.Experiments show that our method performs well while retrieving face from the database according to a user’s input face.Our method can be applied to face recognition,digital entertainment field etc.

Key words: Hausdorff distance, image match, face retrieval, Active Shape Model(ASM)