Abstract | ||
---|---|---|
Based on the equivalence between canonical correlation analysis (CCA) and Fisher linear discriminant analysis (FLDA), two methods for feature extraction of face images are proposed in this paper. In the first approach, the high-dimensional face images are first mapped into the range space of total scatter matrix using principle component analysis (PCA). Then CCA is performed to extract the linear optimal discriminant features without losing Fisher discriminatory information. In the second approach, nonlinear features are extracted using KPCA+CCA which is equivalent to KFDA in nature. The experimental results upon ORL face database indicate that the proposed PCA/KPCA+CCA significantly outperform the traditional Fisherface method. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1007/11539117_11 | ICNC (2) |
Keywords | Field | DocType |
fisher discriminatory information,face image,face recognition,fisher linear discriminant analysis,high-dimensional face image,linear optimal discriminant feature,orl face database,principle component analysis,canonical correlation analysis,proposed pca,linear optimization,feature extraction,scattering matrix | Facial recognition system,Pattern recognition,Canonical correlation,Discriminant,Feature extraction,Artificial intelligence,Fisher information,Linear discriminant analysis,Scatter matrix,Principal component analysis,Mathematics | Conference |
Volume | ISSN | ISBN |
3611 | 0302-9743 | 3-540-28325-0 |
Citations | PageRank | References |
3 | 0.60 | 5 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yunhui He | 1 | 9 | 2.45 |
Li Zhao | 2 | 380 | 27.36 |
Cairong Zou | 3 | 415 | 27.19 |