By shifting the noise into higher-order components, you can discard those components entirely, effectively "cleaning" the dataset before further analysis.
The first step uses a noise covariance matrix (often estimated from dark current or uniform areas of an image) to "whiten" the noise. This makes the noise variance equal in all bands and uncorrelated between bands. mnf encode
The second step performs a standard PCA on the noise-whitened data. This separates the noise from the signal, resulting in a set of components (eigenvectors) where the initial components contain the most signal and the later components contain mostly noise. Why "Encode" with MNF? By shifting the noise into higher-order components, you
