A Declustering Criterion for Feature Extraction in Pattern Recognition
- 1 March 1978
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. C-27 (3), 261-266
- https://doi.org/10.1109/tc.1978.1675083
Abstract
A feature extraction technique based on a new criterion for "declustering" is presented. Declustering occurs when sample vectors from one pattern class form a densely packed point constellation, or cluster, in feature space while vectors from another class do not form a cluster but instead array themselves as outliers. Features chosen to optimize the declustering criterion enhance class separation and are robust over a wide range of measurement statistics.Keywords
This publication has 10 references indexed in Scilit:
- Signal processing for feature extraction and pattern recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Computer Diagnosis of Breast ThermogramsRadiology, 1975
- An Optimal Set of Discriminant VectorsIEEE Transactions on Computers, 1975
- Patterns in pattern recognition: 1968-1974IEEE Transactions on Information Theory, 1974
- Bibliography on estimation of misclassificationIEEE Transactions on Information Theory, 1974
- A Comparison of Seven Techniques for Choosing Subsets of Pattern Recognition PropertiesIEEE Transactions on Computers, 1971
- Nonparametric Bayes-risk estimationIEEE Transactions on Information Theory, 1971
- Application of the Karhunen-Loève Expansion to Feature Selection and OrderingIEEE Transactions on Computers, 1970
- Nonparametric feature selectionIEEE Transactions on Information Theory, 1969
- On the generalized Karhunen-Loeve expansion (Corresp.)IEEE Transactions on Information Theory, 1967