点击次数：

论文类型：期刊论文

发表时间：2009-02-01

发表刊物：Journal of Computational Information Systems

收录刊物：EI、Scopus

卷号：5

期号：1

页面范围：305-310

ISSN号：15539105

摘要：This paper presents a new face recognition method under variable lighting conditions. The proposed method has the ability to factorize a single face image and obtain the illumination invariant facial structure, which is then used for face recognition. Compared with the traditional illumination normalization methods, our method has better edge-preserving ability. In the face recognition step, we present a topology preserving nonlinear mapping method. The key idea is to construct a nonlinear function, which maps the original training space to a feature space. The feature space should have the similar topology structure to the original space, and should be more distinctive. Finally, the results of experiments show that our proposed method is more stable to variable illuminations. Copyright ? 2009 Binary Information Press.