Minimum noise fraction

Minimum noise fraction (MNF) is a transform closely related to PCA, but with an added transformation to whiten any noise present in the data. The resulting transform will be similar to a traditional PCA with the signal- to-noise ratio maximized.

image

This image shows the input Bare Earth Composite, MNF Component 1, and MNF component 15. As MNF orders the components by their signal to noise ratio, MNF 15 is almost completely noise.

Usage

The MNF dialog allows you to select the product and bands for analysis in the same manner as PCA. Two AOIs are required. The first (Area of interest) contains the data from which the data covariances will be computed, and the second (Noise area) contains the data for the noise covariance. Input a name for the output, and click Run MNF to compute the transformation.

Tip

Flat, spectrally homogenous areas work best for input to the noise area. Use the spectral query tool to analyze the spectra of your data to help select an appropriate AOI.

Warning

Although the transformation will be applied dynamically to the layer, keep in mind that it was computed over a specific AOI, and the variance and loadings are only valid over the input AOI. If you are moving to a new geologic regime, it is often necessary and wise to recompute the transformation in the new area instead of relying on the original statistics.