KⅢ模型及其在人脸识别中的应用  

KⅢ model and its application to face recognition

作　　者：张锦[1] 李司铎[1] 李光[2] 

机构地区：[1]湖南大学软件学院,长沙410082 [2]浙江大学工业控制国家重点实验室,杭州310027

出　　处：《计算机工程与应用》2008年第13期245-248,共4页Computer Engineering and Applications

基　　金：国家重点基础研究发展规划(973)(the National Grand Fundamental Research 973 Program of China under Grant No.2004CB720302);国家创新研究群体科学基金(No.60421002);湖南省教育厅资助科研课题(the Research Project of Department of Education of Hunan Province,China under Grant No.07D046)

摘　　要：对一种模拟嗅觉神经系统的仿生模型、KⅢ模型、及其在人脸识别中的应用进行了研究。论文首先分析了KⅢ模型模式识别原理,然后基于人脸识别对模型的模式容量问题进行了研究。使用子图特征组合算法提取人脸特征后,在ORL人脸数据库上的实验结果表明,KⅢ模型的模式容量至少可以达到40,这也更新了之前认为KⅢ模型模式容量约为26的观点。A bionic mode simulating olfactory neural system,KIII model,and its application to face recognition are researched in this paper. Firstly,the pattern recognition principle of KIII model is analyzed. Then,its capability to recognizing different patterns is researched based on face recognition. The face feature is extracted using sub-image feature combination algorithm. According to ORL face database,experimental results show that KIII model can recognize more than 40 patterns. It renews the opinion that KIII model can only recognize about 26 patterns.

关 键 词：嗅觉神经系统 仿生模型 人脸识别 相图 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]