Chang Beom Lee , Do Hyang Kim , Jang Sun Baek , Hyuk Ro Park

The KIPS Transactions:PartB , Vol. 13, No. 4, pp. 393-400, Aug. 2006
10.3745/KIPSTB.2006.13.4.393,   PDF Download:

Abstract

There are many feature extraction methods for face recognition. We need a new method to overcome the small sample problem that the number of feature variables is larger than the sample size for face image data. The paper considers partial least squares(PLS) as a new dimension reduction technique for feature vector. Principal Component Analysis(PCA), a conventional dimension reduction method, selects the components with maximum variability, irrespective of the class information. So, PCA does not necessarily extract features that are important for the discrimination of classes. PLS, on the other hand, constructs the components so that the correlation between the class variable and themselves is maximized. Therefore PLS components are more predictive than PCA components in classification. The experimental results on Manchester and ORL databases shows that PLS is to be preferred over PCA when classification is the goal and dimension reduction is needed.

Statistics
Cite this article