Digital image pattern recognition system using normalized Fourier transform and normalized analytical Fourier-Mellin transform
Rodrigo Vélez-Rábago, Selene Solorza-Calderón, Adina Jordan-Aramburo · AIP conference proceedings · 2016
This work presents an image pattern recognition system invariant to translation, scale and rotation. The system uses the Fourier transform to achieve the invariance to translation and the analytical Forier-Mellin transform for the invariance to scale and rotation. According with the statistical theory of box-plots, the pattern recognition system has a confidence level at least of 95.4%.