Share Email Print

Proceedings Paper

Optical-digital image classifier using a correlation algorithm for invariant feature extraction
Author(s): V. V. Dadeshidze; Anatoly A. Vasiliev
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

An approach to hybrid optical/digital pattern recognition and scene analysis is described which uses a correlation procedure to compare the initial image (object, scene) with its rotated copy. The correlation function of object description obtained via this procedure is shown to possess scale and rotation invariance and to contain information about the object location in the scene being analyzed. The possibility of locating the objects and their classification using the correlation procedure is demonstrated by computer simulation and experiments performed in a coherent optical correlator with a liquid crystal light valve in the Fourier plane.

Paper Details

Date Published: 3 September 1993
PDF: 11 pages
Proc. SPIE 1978, Coherent Measuring and Data Processing Methods and Devices, (3 September 1993); doi: 10.1117/12.155061
Show Author Affiliations
V. V. Dadeshidze, P.N. Lebedev Physical Institute (Russia)
Anatoly A. Vasiliev, P.N. Lebedev Physical Institute (Russia)

Published in SPIE Proceedings Vol. 1978:
Coherent Measuring and Data Processing Methods and Devices
Valery I. Mandrosov, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?