Dimensionality reduction application

4.5.1 Dimensionality reduction application

Perform Dimension reduction of the input image.

Detailed description

Performs dimensionality reduction on input image. PCA,NA-PCA,MAF,ICA methods are available.

Parameters

This section describes in details the parameters available for this application. Table 4.27, page 387 presents a summary of these parameters and the parameters keys to be used in command-line and programming languages. Application key is DimensionalityReduction.


Parameter key	Parameter type	Parameter description

in	Input image	Input Image
out	Output image	Output Image
rescale	Group	Rescale Output.
rescale.outmin	Float	Output min value
rescale.outmax	Float	Output max value
outinv	Output image	Inverse Output Image
method	Choices	Algorithm
method pca	Choice	PCA
method napca	Choice	NA-PCA
method maf	Choice	MAF
method ica	Choice	ICA
method.napca.radiusx	Int	Set the x radius of the sliding window.
method.napca.radiusy	Int	Set the y radius of the sliding window.
method.ica.iter	Int	number of iterations
method.ica.mu	Float	Give the increment weight of W in [0, 1]
nbcomp	Int	Number of Components.
normalize	Boolean	Normalize.

Table 4.27: Parameters table for Dimensionality reduction application.

Input Image The input image to apply dimensionality reduction.

Output Image output image. Components are ordered by decreasing eigenvalues.

Rescale Output.

Output min value: Minimum value of the output image.
Output max value: Maximum value of the output image.

Inverse Output Image reconstruct output image.

Algorithm Selection of the reduction dimension method. Available choices are:

PCA: Principal Component Analysis.
NA-PCA: Noise Adjusted Principal Component Analysis.
- Set the x radius of the sliding window.:
- Set the y radius of the sliding window.:
MAF: Maximum Autocorrelation Factor.
ICA: Independent Component Analysis.
- number of iterations :
- Give the increment weight of W in [0, 1]:

Number of Components. Number of relevant components kept. By default all components are kept.

Normalize. center AND reduce data before Dimensionality reduction.

Example

To run this example in command-line, use the following:

otbcli_DimensionalityReduction -in cupriteSubHsi.tif -out FilterOutput.tif -method pca

To run this example from Python, use the following code snippet:

#!/usr/bin/python

# Import the otb applications package
import otbApplication

# The following line creates an instance of the DimensionalityReduction application
DimensionalityReduction = otbApplication.Registry.CreateApplication("DimensionalityReduction")

# The following lines set all the application parameters:
DimensionalityReduction.SetParameterString("in", "cupriteSubHsi.tif")

DimensionalityReduction.SetParameterString("out", "FilterOutput.tif")

DimensionalityReduction.SetParameterString("method","pca")

# The following line execute the application
DimensionalityReduction.ExecuteAndWriteOutput()

Limitations

Though the inverse transform can be computed, this application only provides the forward transform for now.

Authors

This application has been written by OTB-Team.