← All NMath Code Examples
Imports System
Imports System.Globalization
Imports System.Threading
Imports System.Text
Imports CenterSpace.NMath.Core
Namespace CenterSpace.NMath.Examples.VisualBasic
A .NET example in Visual Basic showing how to use the Discrete Wavelet Transform (DWT) classes.
Module DWTExample
Sub Main()
This section shows to use various configuration settings. All of these settings can
also be set via configuration file or by environment variable. More on this at:
http://www.centerspace.net/blog/nmath/nmath-configuration/
For NMath to continue to work after your evaluation period, you must set your license
key. You will receive a license key after purchase.
More here:
http://www.centerspace.net/blog/nmath/setting-the-nmath-license-key/
NMathConfiguration.LicenseKey = "<key>"
This will start a log file that you can use to ensure that your configuration is
correct. This can be especially useful
for deployment. Please turn this off when you are convinced everything is
working. If you arent sure, please send the resulting log file to
support@centerspace.net. Please note that this directory can be relative or absolute.
NMathConfiguration.LogLocation = "<dir>"
NMath loads native, optimized libraries at runtime. There is a one-time cost to
doing so. To take control of when this happens, use Init(). If your program calls
Init() successfully, then your configuration is definitely correct.
NMathConfiguration.Init()
Console.WriteLine()
DWT of a signal using the Harr wavelet.
Do a simple DWT on a signal
Dim Data As New DoubleVector(12, New RandGenNormal(0.0, 2.0))
Choose wavelet
Dim WaveletInstance As New DoubleWavelet(Wavelet.Wavelets.Harr)
Build DWT object
Dim DWT As New DoubleDWT(WaveletInstance)
Decompose signal with DWT
Dim Approx() As Double
Dim Details() As Double
DWT.DWT(Data.DataBlock.Data, Approx, Details)
Rebuild the signal
Dim Signal() As Double = DWT.IDWT(Approx, Details)
Print results
Console.WriteLine()
Console.WriteLine("A DWT signal decomposition and reconstruction example using the Harr wavelet.")
Console.WriteLine(String.Format("Original Signal: {0}", Data.ToString("#.##")))
Console.WriteLine(String.Format("DWT Approximation: {0}", New DoubleVector(Approx).ToString("#.##")))
Console.WriteLine(String.Format("DWT Details: {0}", New DoubleVector(Details).ToString("#.##")))
Console.WriteLine(String.Format("IDWT Reconstructed signal: {0}", New DoubleVector(Signal).ToString("#.##")))
DWT using a Daubeachies wavelet, then thresholding, and finally reconstucting the signal.
An example to filter a signal using the DWT.
Build random signal data
Data = New DoubleVector(26, New RandGenNormal(1.0, 1.0))
Choose wavelet
WaveletInstance = New DoubleWavelet(Wavelet.Wavelets.D2)
Build DWT object
DWT = New DoubleDWT(Data.DataBlock.Data, WaveletInstance)
Decompose signal with DWT to level 5
DWT.Decompose(3)
Find Universal threshold
Dim LambdaU As Double = DWT.ComputeThreshold(DoubleDWT.ThresholdMethod.Universal, 1)
Threshold all detail levels with lambdaU
DWT.ThresholdAllLevels(DoubleDWT.ThresholdPolicy.Soft, New Double() {LambdaU, LambdaU, LambdaU, LambdaU, LambdaU})
Rebuild signal to level 2
Dim ReconstructedData2() As Double = DWT.Reconstruct(2)
Rebuild the signal to level 1 - the original (filtered) signal.
Dim ReconstructedData1() As Double = DWT.Reconstruct()
Print results
Console.WriteLine()
Console.WriteLine("A DWT signal thresholding and reconstruction example using a Daubeachies wavelet.")
Console.WriteLine(String.Format("Original Signal: "))
Console.WriteLine(" {0}", Data.ToString("#.##"))
Console.WriteLine()
Console.WriteLine(String.Format("IDWT Reconstructed signal: "))
Console.WriteLine(" {0}", New DoubleVector(ReconstructedData1).ToString("#.##"))
Console.WriteLine()
Console.WriteLine("Finished. Press Enter Key.")
Console.Read()
End Sub
End Module
End Namespace
← All NMath Code Examples