COMPLEX WAVELET TRANSFORM IN PATTERN RECOGNITION

MUSOKO, V.; PROCHÁZKA, A.

Abstract

The paper presents basic properties of the complex wavelet transform, the dual tree signal decomposition and features of such a decomposition in the comparison with the wavelet transform using wavelets defined either in the analytical or recurrent forms. This approach to signal and image analysis is studied in the mathematical form and verified for simulated structures as well. The final part of the paper presents the use of this method in feature extraction, pattern recognition and classification.

Coresponding author e-mail: Victor[dot]Musoko[at]vscht[dot]cz

Session: Modelling, Simulation, and Identification of Processes