Principal component analysis

A data transformation method that rotates the axes of the input bands to a new multivariate attribute space in which the axes are uncorrelated. The main reason to transform the data in a principal component analysis is to compress data by eliminating redundancy.

» Remote Sensing and GIS glossary