Main Content

Empirical wavelet transform

returns
the multiresolution analysis (MRA) components corresponding to the empirical wavelet
transform (EWT) of `mra`

= ewt(`x`

)`x`

. Use `ewt`

to decompose
signals using an adaptable wavelet subdivision scheme that automatically determines the
empirical wavelet and scaling filters and preserves energy.

By default, the number of empirical wavelet filters is automatically determined by
identifying peaks in a multitaper power spectral estimate of
`x`

.

`[___] = ewt(___,`

specifies additional options using name-value pair arguments. These arguments can be added
to any of the previous input syntaxes. For example, `Name,Value`

)`'MaxNumPeaks',5`

specifies a maximum of five peaks used to determine the EWT filter passbands.

`ewt(___)`

with no output arguments plots the original
signal with the empirical wavelet MRA in the same figure. For complex-valued data, the
real part is plotted in the first color in the MATLAB^{®} color order matrix and the imaginary part is plotted in the second
color.

[1] Gilles, Jérôme. “Empirical Wavelet
Transform.” *IEEE Transactions on Signal Processing* 61, no. 16 (August
2013): 3999–4010. https://doi.org/10.1109/TSP.2013.2265222.

[2] Gilles, Jérôme, Giang Tran, and
Stanley Osher. “2D Empirical Transforms. Wavelets, Ridgelets, and Curvelets Revisited.”
*SIAM Journal on Imaging Sciences* 7, no. 1 (January 2014): 157–86.
https://doi.org/10.1137/130923774.

[3] Gilles, Jérôme, and Kathryn Heal.
“A Parameterless Scale-Space Approach to Find Meaningful Modes in Histograms — Application to
Image and Spectrum Segmentation.” *International Journal of Wavelets,
Multiresolution and Information Processing* 12, no. 06 (November 2014): 1450044.
https://doi.org/10.1142/S0219691314500441.