1 September 2008 Optical pattern recognition using multiple phase-shifted-reference fringe-adjusted joint transform correlation
Author Affiliations +
Optical Engineering, 47(9), 097204 (2008). doi:10.1117/1.2981514
Abstract
We propose a novel optical pattern recognition system using multiple phase-shifted-reference fringe-adjusted joint transform correlation (MRFJTC) techniques. The MRFJTC algorithm can efficiently detect an object of interest in the input scene by producing a highly distinctive correlation peak while rejecting any and all nontarget objects in a complex background. The simple architecture of the proposed system can simultaneously recognize multiple targets of the reference class when present. The recognition performance is fast, automatic, and invariant to noise and distortions.
Mohammed Nazrul Islam, Vijayan K. Asari, Mohammad A. Karim, "Optical pattern recognition using multiple phase-shifted-reference fringe-adjusted joint transform correlation," Optical Engineering 47(9), 097204 (1 September 2008). http://dx.doi.org/10.1117/1.2981514
JOURNAL ARTICLE
9 PAGES


SHARE
KEYWORDS
Optical pattern recognition

Target recognition

Target detection

Pattern recognition

Joint transforms

Optical engineering

Phase shifts

Back to Top