基于隶属度的人脸图像特征抽取和识别  

Human Facial Image Feature Extraction and Recognition Based on Degree of Membership

作　　者：黄修武[1] 杨静宇[1] 郭跃飞[1] 

机构地区：[1]南京理工大学计算机科学与工程系,南京210094

出　　处：《电子学报》1998年第5期89-92,共4页Acta Electronica Sinica

基　　金：国家自然科学基金!69672013;国家教委博士点基金

摘　　要：本文首先从代数理论和模糊数学导出一种新概念：矢量隶属函数，由此构造图像的隶属函数，并证明它具有一些不变性，提出一种基于隶属函数的图像特征抽取和识别方法，并应用于人脸图像识别问题．实验结果表明，隶属函数具有良好的识别分离能力．A new concept-function of vector membership from algebraic theory and fuzzy math is derived.The function of image membership is constructed and a feature extraction and recognition method is presented based on the function of image membership. Finally,this method is used to solve the recognition problem of human facial images. Experimental results have shown that the membership has effective ability of recognition and distinction.

关 键 词：隶属度 代数特征抽取 图像识别 人脸图像特征 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]