A Novel Regularized Fisher Discriminant Method for Face Recognition Based on Subspace and Rank Lifting Scheme

The null space N(St) of total scatter matrix St contains no useful information for pattern classification. So, discarding the null space N(St) results in dimensionality reduction without loss discriminant power. Combining this subspace technique with proposed rank lifting scheme, a new regularized Fisher discriminant (SL-RFD) method is developed to deal with the small sample size (S3) problem in face recognition. Two public available databases, namely FERET and CMU PIE databases, are exploited to evaluate the proposed algorithm. Comparing with existing LDA-based methods in solving the S3 problem, the proposed SL-RFD method gives the best performance. Keywords. Face recognition, linear discriminant analysis, small sample size problem, Regularized method, Null space.

• Type: Conference
• Year: 2005
• Where: ACII
• Authors: Wen-Sheng Chen, Pong Chi Yuen, Jian Huang, Jian-Huang Lai, Jianliang Tang
• Keywords: ACII 2005, Applied Computing, Discriminant, Null Space, Small Sample Size