DocumentCode :
1947088
Title :
Comparison of several learning subspace methods for classification
Author :
Taur, J.S. ; Kung, S.Y.
Author_Institution :
Dept. of Electr. Eng., Princeton Univ., NJ, USA
fYear :
1991
fDate :
14-17 Apr 1991
Firstpage :
1069
Abstract :
Several competition-based methods for classification are compared. Special attention is paid to subspace methods which are based on computing the projections of the patterns on the principal component vectors of the correlation matrices that span the pattern subspaces. A decision learning rule which updates the correlation matrices can be used to adjust the class boundary and improve the performance of the classification. A learning subspace method is proposed, and some other classification methods are reviewed. In this comparison, all of the methods are applied to a texture classification problem and the performance results are presented
Keywords :
learning systems; neural nets; pattern recognition; state-space methods; class boundary; classification methods; correlation matrices; decision learning rule; learning subspace methods; pattern classification; pattern subspaces; performance; principal component vectors; texture classification; Classification algorithms; Labeling; Mean square error methods; Statistics;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
ISSN :
1520-6149
Print_ISBN :
0-7803-0003-3
Type :
conf
DOI :
10.1109/ICASSP.1991.150540
Filename :
150540
Link To Document :
بازگشت