Abstract: An attribute means clustering binary tree is presented in this paper. The binary tree is extends naturally and turns to be a supervised classification method. The ORL database is used to evaluate the proposed method. The performance of the attribute means clustering binary tree used in face recognition is compared with the standard eigenface approach and improves their performance much in the experiment.

Key words: pattern recognition, attribute means clustering, learn network, binary tree, eigenface

摘要： 在无监督的属性聚类网络的基础上，提出了一种二叉树分类方法。此二叉树自然地在无监督聚类的基础上扩展开来，成为一有监督的分类方法。用ORL人脸数据库做了测试，同标准的特征脸(eigenface)方法相比，识别率得到了较大的提高。

关键词: 模式识别, 属性均值聚类, 学习网络, 二叉树, 特征脸