Advertisement

Feature Reduction

  • John A. Richards
Chapter

Abstract

Many remote sensing instruments record more channels or bands of data than are actually needed for most applications. As an example, even though the Hyperion sensor on EO-1 produces 220 channels of image data over the wavelength range 0.4–2.4 μm, it is unlikely that channels beyond about 1.0 μm would be relevant for water studies, unless the water were especially turbid. Furthermore, unless the actual reflectance spectrum of the water was essential for the task at hand, it may not even be necessary to use all the contiguous bands recorded in the range 0.4–1.0 μm; instead, a representative subset may be sufficient in most cases.

Keywords

Feature Reduction Canonical Analysis Pairwise Divergence Spectral Classis Maximum Likelihood Classification 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.ANU College of Engineering and Computer ScienceAustralian National UniversityCanberraAustralia

Personalised recommendations