Individual bands of multi-spectral remote sensing data are referred
to as *features* in the field of pattern recognition. We may
then define a *feature vector* as the *n*-dimensional
vector associated with the brightness values in *n* multi-
spectral bands for a pixel. The vector space associated with
these vectors is often called the *feature space* or *spectral
space*. Statistical data reduction techniques, such as
** Principal Component
Analysis** can be used to reduce the dimensionality
of feature space.