Nonorthogonal Projections for Feature Extraction in Pattern Recognition
- 1 May 1970
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. C-19 (5) , 447-452
- https://doi.org/10.1109/T-C.1970.222943
Abstract
It is known that R linearly separable classes of multidimensional pattern vectors can always be represented in a feature space of at most R dimensions. An approach is developed which can frequently be used to find a nonorthogonal transformation to project the patterns into a feature space of considerably lower dimensionality. Examples involving classification of handwritten and printed digits are used to illustrate the technique.Keywords
This publication has 10 references indexed in Scilit:
- Randomly Generated Nonlinear Transformations for Pattern RecognitionIEEE Transactions on Systems Science and Cybernetics, 1969
- Nonparametric feature selectionIEEE Transactions on Information Theory, 1969
- A Nonlinear Mapping for Data Structure AnalysisIEEE Transactions on Computers, 1969
- Feature extraction: A surveyProceedings of the IEEE, 1969
- On mode estimation in pattern recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1968
- Pattern Recognition from Satellite AltitudesIEEE Transactions on Systems Science and Cybernetics, 1968
- Mapping multidimensional space to one dimension for computer output displayPublished by Association for Computing Machinery (ACM) ,1968
- An Algorithm for Linear Inequalities and its ApplicationsIEEE Transactions on Electronic Computers, 1965
- Data for Character Recognition StudiesIEEE Transactions on Electronic Computers, 1963
- An Analog Method for Character RecognitionIEEE Transactions on Electronic Computers, 1961