基于虚拟样本扩张法的单样本人脸识别算法研究

随着人脸识别技术的不断发展,单样本人脸识别已成为当今的一个热点。针对单样本人脸识别问题,提出了一种基于虚拟样本扩展的人脸识别方法,为给定的单训练样本增加虚拟图像,以增强单训练样本的分类信息,并对原样本及其虚拟样本进行特征变换,划分得到更多的子图像,利用二维主成分分析(2DPCA)实现特征抽取,一定程度上减轻了人脸的表情、姿态、光照等因素对识别效果的影响,提高了识别率。提出的方法分别在ORL及FERET两大人脸数据库上得到了验证。

Virtual Sample Generating for Face Recognition from A Single Training Sample per Person

Usually, we assume that there are multiple samples per person for feature extraction in many face recognition methods. In many practical face recognition applications such as law enhancement, e-passport, and ID card identification, however, this assumption may not hold as there is only a single sample per person. To address this problem, we propose a novel method in this paper that generating multiple samples with sample give by using virtual sample generating method so as to adding classes of each face, partitioning them into multiple sub-patches, using 2DPCA method to complete feature extraction . Experiment results on two widely used face databases are presented to demonstrate the efficacy of the proposed approach.