基于表示的简单快速人脸识别方法

提出了一种简单且快速的基于表示的人脸识别方法,并给出其内在的原理.首先,该方法从每一个类的测试集中,只选择最近的训练样本;然后,将测试样本表示为所有选定训练样本的线性组合;最后,由欧氏度量进行决策,通过修改测试样本与训练样本之间的邻居关系,达到更好的识别效果.ORL及Yale人脸数据库的实验结果表明,所提方法可对测试样本进行高准确度的分类,比最近邻分类法分类更准确,相比几种常用的特征空间方法具有更高的识别率.

A Simple and Fast Face Recognition Method Based on Representation

In this paper, a very simple and fast face recognition method (RSF) was proposed and its potential rationale was presented. This method ?rst selects only the nearest training sample, of the test sample, from every class and then expresses the test sample as a linear combination of all the selected training samples. Using the expression result, the proposed method can classify the testing sample with a high accuracy. Our method can classify more accurately than the nearest neighbor classi?cation method (NNCM). Experiments on databases of ORL and Yale show that the classi?cation accuracy obtained using our method was usually greater than that obtained using NNCM. It can also obtain a better performance than the nearest feature space method.