Inverse transform ContinuousWaveletData from the forward transform:

The quality of the reconstruction depends on the number of octaves and voices:

Inverse transform modified ContinuousWaveletData:

Plot the inverse transform of original and modified coefficients:

Inverse transform selected octaves and voices only:

Inverse transform only the {2,5} coefficient:

Inverse transform the first octave {1,_}, setting other coefficients to zero:

Inverse transform an explicitly constructed ContinuousWaveletData object:

Unspecified coefficients are taken to be zero:

Specify a different wavelet to use in the inverse transform:

By default, the wavelet used in the forward transform is chosen:

By default, Method option "LeastSquares" is used for data less than length 512:

By default, Method option "DeltaFunction" is used for data greater than length 512:

Method option "LeastSquares" performs an exact inverse transform:

Method option "DeltaFunction" performs an approximate inverse transform:

Compare efficiency and accuracy of the two methods:

For large data, "LeastSquares" is slow:

Compare with the original data:

Use "DeltaFunction" for large data:

Compare with the original data:

Filter one scale or frequency from a signal:

Identify separate signal components on a scalogram:

Remove the feature at small scales {3|4|5,_}:

Synthesize filtered data:

Filter data with both time- and scale-dependent features:

Identify signal components as a function of scale and time:

Excise transient feature using a step filter:

Show altered scalogram and synthesized filtered data: