Abstract

A recently proposed position, rotation, and scale-invariant optical pattern recognition system using a holographic logpolar mapping filter is analyzed with respect to resolution and accuracy. The limits are set by the resolution of the mapping filter, which must have a bandwidth proportional to the square of the input bandwidth. With 1800 × 1800 pixels in the computer-generated mapping filter, we get a maximum input field of 24 × 24 pixels and an angular resolution of 4.8°. Determination of the position of the input object is possible with at best 50% relative accuracy, and, in practice, multiobject capability is highly questionable. Numerical simulations and experimental evidence support the resolution limits found.

© 1988 Optical Society of America

Computerized design and generation of space-variant holographic filters. 1: System design considerations and applications of space-variant filters to image processing

P. Ambs, Y. Fainman, S. H. Lee, and J. Gresser
Appl. Opt. 27(22) 4753-4760 (1988)

Real-time deformation invariant optical pattern recognition using coordinate transformations

David Casasent, Shao-Feng Xia, Andrew J. Lee, and Jian-Zhong Song
Appl. Opt. 26(5) 938-942 (1987)

Coordinate-transformed filter for shift-invariant and scale-invariant pattern recognition

Dan Cojoc, M. Teresa Molina, Javier García, and Carlos Ferreira
Appl. Opt. 36(20) 4812-4815 (1997)

Shift-and-scale-invariant pattern recognition using an elliptic coordinate-transformed phase-only filter

Javier García, Tomasz Szoplik, and Carlos Ferreira
Appl. Opt. 31(23) 4823-4828 (1992)

Circular harmonic phase filters for efficient rotation-invariant pattern recognition

Joseph Rosen and Joseph Shamir
Appl. Opt. 27(14) 2895-2899 (1988)