Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Constructing descriptive and discriminative nonlinear features - Rayleigh coefficients in kernel feature spaces

: Mika, S.; Ratsch, G.; Weston, J.; Schölkopf, B.; Smola, A.; Müller, K.-R.


IEEE Transactions on Pattern Analysis and Machine Intelligence 25 (2003), No.5, pp.623-628
ISSN: 0162-8828
Journal Article
Fraunhofer FIRST ()

We incorporate prior knowledge to construct nonlinear algorithms for invariant feature extraction and discrimination. Employing a unified framework in terms of a nonlinearized variant of the Rayleigh coefficient, we propose nonlinear generalizations of Fisher's discriminant and oriented PCA using support vector kernel functions. Extensive simulations show the utility of our approach.