14 August 2001 Nonlinear binary correlations for rotation-invariant pattern recognition using circular harmonic decomposition
Author Affiliations +
Proceedings Volume 4419, 4th Iberoamerican Meeting on Optics and 7th Latin American Meeting on Optics, Lasers, and Their Applications; (2001) https://doi.org/10.1117/12.437126
Event: IV Iberoamerican Meeting of Optics and the VII Latin American Meeting of Optics, Lasers and Their Applications, 2001, Tandil, Argentina
Abstract
We apply the sliced orthogonal nonlinear generalized (SONG) correlation to rotation invariant pattern recognition using circular harmonic decompositions. The images are decomposed into disjoint binary slices to correlate each of them with one object binary slice circular harmonic component. To obtain the final Rotation Invariant SONG correlation output, we add the correlations from each slice. In addition, in order to improve discrimination capability, the method avoids the time-consuming process of finding proper center for the circular harmonic decompositions.
© (2001) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Pascuala Garcia-Martinez, Pascuala Garcia-Martinez, Henri H. Arsenault, Henri H. Arsenault, Carlos Ferreira, Carlos Ferreira, "Nonlinear binary correlations for rotation-invariant pattern recognition using circular harmonic decomposition", Proc. SPIE 4419, 4th Iberoamerican Meeting on Optics and 7th Latin American Meeting on Optics, Lasers, and Their Applications, (14 August 2001); doi: 10.1117/12.437126; https://doi.org/10.1117/12.437126
PROCEEDINGS
4 PAGES


SHARE
Back to Top