Fast hyperspectral feature reduction using piecewise constant function approximations

Are Charles Jensen

doi:10.1109/LGRS.2007.896331

Fast hyperspectral feature reduction using piecewise constant function approximations

2007-10-01

34

PRE

AI

A

Are Charles Jensen *

DOI：10.1109/LGRS.2007.896331

Original

Original request for help

Save

Abstract

En 中文

The high number of spectral bands that are obtained from hyperspectral sensors, combined with the often limited ground truth, solicits some kind of feature reduction when attempting supervised classification. This letter demonstrates that an optimal constant function representation of hyperspectral curves in the mean square sense is capable of representing the data sufficiently to outperform, or match, other feature reduction methods such as principal components transform, sequential forward selection, and decision boundary feature extraction for classification purposes on all of the four hyperspectral data sets that we have tested. The simple averaging of spectral bands makes the resulting features directly interpretable in a physical sense. Using an efficient dynamic programming algorithm, the proposed method can be considered fast.

Keywords:

feature extraction

pattern classification

remote sensing

Fast hyperspectral feature reduction using piecewise constant function approximations

Abstract

Journal

Organization