The Community for Technology Leaders
Pattern Recognition, International Conference on (2000)
Barcelona, Spain
Sept. 3, 2000 to Sept. 8, 2000
ISBN: 0-7695-0750-6
pp: 2398
Robert P.W. Duin , Delft University of Technology
Marco Loog , Delft University of Technology and Philips Research Laboratories
R. Haeb-Umbach , Philips Research Laboratories
ABSTRACT
The traditional way to find a linear solution to the feature extraction problem is based on the maximization of the class-between scatter over the class-within scatter (Fisher mapping). For the multi-class problem this is, however, sub-optimal due to class conjunctions, even for the simple situation of normal distributed classes with identical covariance matrices. We propose a novel, equally fast method, based on nonlinear PCA. Although still sub-optimal, it may avoid the class conjunction. The proposed method is experimentally compared with Fisher mapping and with a neural network based approach to nonlinear PCA. It appears to outperform both methods, the first one even in a dramatic way.
INDEX TERMS
CITATION

M. Loog, R. P. Duin and R. Haeb-Umbach, "Multi-Class Linear Feature Extraction by Nonlinear PCA," Pattern Recognition, International Conference on(ICPR), Barcelona, Spain, 2000, pp. 2398.
doi:10.1109/ICPR.2000.906096
95 ms
(Ver 3.3 (11022016))