Computes the multi-level undecimated wavelet transform (UWT) of signal. This VI returns the approximation coefficients at the largest level and the detail coefficients at all levels for a 1D signal input and returns the approximation coefficients and the detail coefficients at all levels for a 2D signal input. The approximation coefficients and the detail coefficients at all levels are the same size as signal. The results of the UWT have the translation invariant property, which is helpful in robust feature extraction and pattern recognition. Wire data to the signal input to determine the polymorphic instance to use or manually select the instance.


icon

WA Undecimated Wavelet Transform Details

The following illustration shows an example of a three-level UWT, where you set the levels input of this VI to 3. The length of the input signal is 16 points.

where the subscript indicates the up-sampling by a factor of m. Using information in the previous illustration, you can see that the UWT coef output contains the approximation coefficients of the largest level and the detail coefficients of each level.

In addition, compared to the discrete wavelet transform (DWT), the UWT does not have sub-sampling. Therefore, the output of each level has the same length as the input signal.

Undecimated discrete wavelet transform is translation-invariant. In other words, when the input signal shifts with certain taps, the output coefficients shift by the same taps, which is important for robust feature extractions, such as peak detection.

Refer to A Wavelet Tour of Signal Processing for more information about the UWT.

Examples

Refer to the following VIs for examples of using the WA Undecimated Wavelet Transform VI:

  • High-order Discontinuity Detection VI: labview\examples\Wavelet Analysis\WAGettingStarted
  • Undecimated Image Decomposition and Reconstruction (UWT) VI: labview\examples\Wavelet Analysis\WAGettingStarted
  • Undecimated Signal Decomposition and Reconstruction (UWT) VI: labview\examples\Wavelet Analysis\WAGettingStarted