Share Email Print

Proceedings Paper

Color Image Correlation: Principle And Application
Author(s): E. Badique; Y. Komiya; N. Ohyama; J. Tsujiuchi; T. Honda
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Despite evidences from human vision research (1) that spatial and color characteristics of images are strongly related, the use of color for recognition purposes has received little attention. However,in the framework of the development of increasingly versatile machine vision systems, the introduction of the color feature seems to be a rather natural step. A color approach is obviously not always possible and justified and therefore we are interested in cases where color images are at hand and where recognition accuracy is more important than processing speed. In this work we extend the concepts introduced earlier (2) andpropose to find an optimal generalized color plane in which to carry out the correlation operation. The optimality is defined in terms of the maximization of the peak-to-deviation ratio in the correlation plane. As an application we consider the retrieval of the mucosal surface structure from stereo pairs taken through an endoscope.

Paper Details

Date Published: 1 January 1987
PDF: 2 pages
Proc. SPIE 0813, Optics and the Information Age, (1 January 1987); doi: 10.1117/12.967218
Show Author Affiliations
E. Badique, Tokyo Institute of Technology (Japan)
Y. Komiya, Tokyo Institute of Technology (Japan)
N. Ohyama, Tokyo Institute of Technology (Japan)
J. Tsujiuchi, Tokyo Institute of Technology (Japan)
T. Honda, Tokyo Institute of Technology (Japan)

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

© SPIE. Terms of Use
Back to Top