A new supervised image classifier architecture based on multiresolution wavelet network including a fuzzy decision support system

Abstract

The problem of image classification remains to be a major challenge to the computer vision community. In this paper, we propose a new classifier architecture based on multiresolution wavelet network learnt by fast wavelet transform including a fuzzy decision support system (FWN-FDSS). The proposed classifier has many advantages compared to other ones. It is characterized by its new method of computing similarity distances and his way of decision-making which operates a human reasoning mode. Comparisons with other classifiers are presented and discussed. Obtained results have shown that the new classifier performs better than previously established ones.