Method of feature extraction using combination of wavelet and fourier transformation

摘要

The present invention is a method of removing at least one feature from a signal by receiving the signal in the time domain; identifying at least one frequency to be removed; identifying a number of resolution levels to which the signal will be reduced using wavelet transformation; computing at least one wavelet difference coefficient for each resolution level and at least one wavelet average coefficient for the lowest resolution level; creating a projection of each at least one frequency onto the lowest resolution level using Fourier transformation; subtracting the projection from the corresponding at least one wavelet average coefficient; and adding the results of the last step to the at least one wavelet difference coefficient.