Feature Extraction Using Problem Localization
- 1 May 1982
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. PAMI-4 (3) , 323-326
- https://doi.org/10.1109/TPAMI.1982.4767252
Abstract
Feature extraction is considered as a mean-quare estimation of the Bayes risk vector. The problem is simplified by partitioning the distribution space into local subregions and performing a linear estimation in each subregion. A modified clustering algorithm is used to fimd the partitioning which minimizes the mean-square error.Keywords
This publication has 4 references indexed in Scilit:
- A class of feature extraction criteria and its relation to the Bayes risk estimateIEEE Transactions on Information Theory, 1980
- Nonlinear feature extraction with a general criterion functionIEEE Transactions on Information Theory, 1978
- Generalized Clustering for Problem LocalizationIEEE Transactions on Computers, 1978
- The optimum nonlinear features for a scatter criterion in discriminant analysisIEEE Transactions on Information Theory, 1977