图象隶属度及其在人脸识别中的应用  

DEGREE OF MEMBERSHIP OF IMAGE AND ITS APPLICATION IN HUMAN FACIAL RECOGNITION

作　　者：黄修武[1] 杨静宇[1] 范伯南[1] 郭跃飞[1] 

机构地区：[1]南京理工大学信息学院计算机科学系

出　　处：《计算机研究与发展》1998年第11期1014-1019,共6页Journal of Computer Research and Development

基　　金：国家自然科学基金;博士点基金

摘　　要：文中首先从代数理论和模糊数学出发，讨论向量空间中任一子空间对任一向量的接纳程度，将其定义为向量隶属度，并证明它具有一些不变性；然后将它推广应用于图象识别中，构造图象隶属度，人脸识别实验结果表明。Starting from algebraic theory and fuzzy mathematics,the degree of any vector subspace accepting any vector is discussed in the paper, which is defined as the degree of membership of vector. It is proved to have some properties of algebraic invariance and then it is applied to image recognition for constructing the degree of membership of image. Experimental results of human facial image recognition show that the degree of membership of image has the good ability of separation and recognition.

关 键 词：图象识别 人脸识别 模式识别 图象隶属度 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]