Improving the eigenphase method for face recognition

IEICE Electronics Express

Jesus Olivares-Mercado²  Karina Toscano-Medina²  Mariko Nakano-Miyatake²  Haruhisa Takahashi¹  Hector Perez-Meana²  Kazuhiro Hotta¹ 

[1] The University of Electro-Communications;National Polytechnic Institute of Mexico

关键词: eigenphases;    GMM;    face recognition and verification;    phase spectrum;   

DOI  :  10.1587/elex.6.1112

学科分类：电子、光学、磁材料

来源: Denshi Jouhou Tsuushin Gakkai

【 摘 要 】

References(7)Cited-By(2)This paper proposes an improvement to the Eigenphases method, in which the image is normalized to reduce the illumination and facial expression effects and the Principal Components Analysis (PCA) is used for feature extraction, while the Gaussian Mixture Model (GMM)is used to improve the performance of classification stage. An important advantage of GMM is that this system is trained without supervisor and constructs an independent model for each user. The proposed method is evaluated using the ”AR Face Database”, which includes the face images of 120 subjects (65 males and 55 females). Evaluation results show that the proposed method provides better performance than the original eigenphases method.

【 授权许可】

Unknown