Title :
Fast spectral algorithms for invariant pattern recognition and image matching based on modular invariants
Author :
Labunets, E. ; Labunets, V. ; Assonov, M. ; Lenz, Reiner
Author_Institution :
Ural State Tech. Univ., Ekaterinburg, Russia
Abstract :
We propose a new algorithm, which needs almost no multiplication. We propose using modular arithmetic of the finite Galois field GF(Q) to map geometrical moments calculation to a fast Fourier-Mellin-Galois transform, which reduces the computational complexity from 𝒪(N4) to 𝒪(N2 log2 N). We illustrate the performance of the method by some classification and matching experiments
Keywords :
Galois fields; computational complexity; digital arithmetic; fast Fourier transforms; feature extraction; image classification; pattern recognition; spectral analysis; computational complexity reduction; experiments; fast Fourier-Mellin-Galois transform; fast spectral algorithms; feature extraction; finite Galois field; geometrical moments calculation; image classification; image matching; invariant pattern recognition; modular arithmetic; modular invariants; performance; Arithmetic; Fast Fourier transforms; Feature extraction; Galois fields; Image classification; Image matching; Image recognition; Layout; Pattern recognition; Polynomials;
Conference_Titel :
Image Processing, 1996. Proceedings., International Conference on
Conference_Location :
Lausanne
Print_ISBN :
0-7803-3259-8
DOI :
10.1109/ICIP.1996.560568
