This tool can be used to perform an inverse principal component
analysis (PCA) on a set of PCA component images. The component
images must have been created using the
** Principal Component
Analysis** tool. PCA is a type of data
transformation that is used with multi-dimensional data,
such as that provided by multi-spectral remotely sensed
imagery. PCA is used for numerous applications including data
reduction, change detection, and noise reduction. When used as a
noise reduction technique, an inverse PCA is performed, leaving out
one or more of the high-order PCA components, which account for
most of the noise variance in the original data set. The decision
of how many components to leave out of the inverse transformation
should be based on an analysis of the PCA output, considering the
amount of explained variance in each PCA component.

The inverse PCA transformation is based on the eigenvectors and
eigenvalues contained in the metadata of the component image header
files (.dep). These entries are written into the header files by the
** Principal Component
Analysis** tool and must not have been
modified for the inverse transformation to work properly.

The following is an example of a Python script using this tool:

```
wd = pluginHost.getWorkingDirectory()
```

# You may have multiple input files but they must

# be separated by semicolons in the string.

inputFiles = wd + "input1.dep" + ";" + wd + "input2.dep" + ";" + wd + "input3.dep"

outputSuffix = "invPCA"

args = [inputFiles, outputSuffix]

pluginHost.runPlugin("InversePrincipalComponentAnalysis", args, False)

This is a Groovy script also using this tool:

```
def wd = pluginHost.getWorkingDirectory()
```

// You may have multiple input files but they must

// be separated by semicolons in the string.

def inputFiles = wd + "input1.dep" + ";" + wd + "input2.dep" + ";" + wd + "input3.dep"

def outputSuffix = "invPCA"

String[] args = [inputFiles, outputSuffix]

pluginHost.runPlugin("InversePrincipalComponentAnalysis", args, false)

- John Lindsay (2012), email: jlindsay@uoguelph.ca

- Jensen, J.R. 2005. Introductory digital image processing: A remote sensing perspective, 3rd Ed. Prentice Hall series in Geographic Information Science, Upper Saddle River, N.J., pp. 526.