Share Email Print

Proceedings Paper

Single-channel nonlinear processor for color optical pattern recognition
Author(s): Pascuala Garcia-Martinez; Joaquin Oton; Jose J. Valles; Henri H. Arsenault
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

We introduce a single channel color nonlinear pattern recognition method. Usually, for color pattern recognition methods, the color are processed and finally its correlation outputs are combined. This is the definition of the common multichannel method. We take the input and reference scenes and we decompose them into their red, green and blue channels to obtain an amplitude and phase distribution. In this paper we study the correlations for highly noisy input iamges. These results are obtained by means of the sliced orthogonal nonlinear generalized correlation which detects the target with high discrimination capability in cases where other methods do not succeed.

Paper Details

Date Published: 23 October 2003
PDF: 9 pages
Proc. SPIE 5202, Optical Information Systems, (23 October 2003); doi: 10.1117/12.504720
Show Author Affiliations
Pascuala Garcia-Martinez, Univ. de Valencia (Spain)
Joaquin Oton, Univ. de Valencia (Spain)
Jose J. Valles, Univ. de Valencia (Spain)
Henri H. Arsenault, COPL, Univ. Laval (Canada)

Published in SPIE Proceedings Vol. 5202:
Optical Information Systems
Bahram Javidi; Demetri Psaltis, Editor(s)

© SPIE. Terms of Use
Back to Top