Title :
Multiresolution using principal component analysis
Author :
Brennan, Vic ; Principe, Jose
Author_Institution :
Dept. of Electr. Eng., Florida Univ., Gainesville, FL, USA
Abstract :
This paper proposes principal component analysis (PCA) to find adaptive bases for multiresolution. An input image is decomposed into components (compressed images) which are uncorrelated and have maximum l 2 energy. With only minor modification, a single layer linear network using the generalized Hebbian algorithm (GHA) is used for multiresolution PCA. The decomposition has been successfully applied to face classification. Good results with biological signals have also been reported
Keywords :
Hebbian learning; feature extraction; image classification; image coding; image representation; image resolution; principal component analysis; PCA; biological signals; compressed images; face classification; feature extraction; generalized Hebbian algorithm; input image decomposition; multiresolution; multiresolution PCA; principal component analysis; signal dependent representations; single layer linear network; Eigenvalues and eigenfunctions; Energy resolution; Equations; Image databases; Matrix decomposition; Neural engineering; Principal component analysis; Scattering; Signal resolution; Symmetric matrices;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2000. ICASSP '00. Proceedings. 2000 IEEE International Conference on
Conference_Location :
Istanbul
Print_ISBN :
0-7803-6293-4
DOI :
10.1109/ICASSP.2000.860149
