Share Email Print

Proceedings Paper

Selection of image features for distribution-map classifiers
Author(s): Tin Kam Ho
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

A useful metric for pattern classification can be derived from a series of one-dimensional maps of empirical class-conditional distributions. To achieve maximum classification accuracy, the distributions need to be projected into subspaces where they are locally dense and the classes are at least partially separable. We describe a method where useful projections are obtained automatically by a sequence of cluster and linear discriminant analyses. Each projection can be considered as a feature that contributes to the elimination of unresolved ambiguities. We discuss conditions under which the method can achieve maximum accuracy. In an experiment of applying the method to machine-printed character images, we show that the method yields a classifier that has very low error rate, and can be improved with additional training data.

Paper Details

Date Published: 30 March 1995
PDF: 10 pages
Proc. SPIE 2422, Document Recognition II, (30 March 1995); doi: 10.1117/12.205813
Show Author Affiliations
Tin Kam Ho, AT&T Bell Labs. (United States)

Published in SPIE Proceedings Vol. 2422:
Document Recognition II
Luc M. Vincent; Henry S. Baird, Editor(s)

© SPIE. Terms of Use
Back to Top