计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (13): 245-248.

• 工程与应用 • 上一篇    

KIII模型及其在人脸识别中的应用

张 锦1,2,李司铎1,李 光2   

  1. 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
  • 收稿日期:2007-12-26 修回日期:2008-02-21 出版日期:2008-05-01 发布日期:2008-05-01
  • 通讯作者: 张 锦

KIII model and its application to face recognition

ZHANG Jin1,2,LI Si-duo1,LI Guang2   

  1. 1. School of Software,Hunan University,Changsha 410082,China
    2. National Laboratory of Industrial Control Technology,Zhejiang University,Hangzhou 310027,China
  • Received:2007-12-26 Revised:2008-02-21 Online:2008-05-01 Published:2008-05-01
  • Contact: ZHANG Jin

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。

关键词: 嗅觉神经系统, 仿生模型, 人脸识别, 相图

Abstract: 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.

Key words: olfactory neural system, bionic model, face recognition, phase diagram