Share Email Print

Proceedings Paper

Component Pattern Analysis Of Multiband Images
Author(s): S. Kawata; K. Sasaki; S. Minami
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Various computerized image-processing methods have been developed for analyzing spatial patterns of specific components in given images. In pattern recognition studies, patterns are classified into different components in the feature space by comparing component features, for example, shape, size, density, histogram, direction, etc. However, it often happens that the components can not be classified in the feature hyperspace because of their mutual dependency in spatial domain. We calim that, in such a case we should use spectral (optical frequency) information of components. Different materials must have different spectroscopic responses to electromagnetic waves of a certain energy band.

Paper Details

Date Published: 1 January 1987
PDF: 2 pages
Proc. SPIE 0813, Optics and the Information Age, (1 January 1987); doi: 10.1117/12.967304
Show Author Affiliations
S. Kawata, Osaka University (Japan)
K. Sasaki, Osaka University (Japan)
S. Minami, Osaka University (Japan)

Published in SPIE Proceedings Vol. 0813:
Optics and the Information Age
Henri H. Arsenault, Editor(s)

© SPIE. Terms of Use
Back to Top