ABSTRACT
In Chapter 5, the fundamental concepts and basics of feature extraction were introduced and summarized. Numerous feature extraction methods have been developed in the past decades, and these methods can be broadly divided into spatial and spectral approaches associated with supervised, semi-supervised, and unsupervised techniques by taking the working principles (i.e., whether the labeled data are required or not) into account. In this chapter, a suite of feature extraction algorithms with the aid of statistical and decision science techniques are summarized for use in remote sensing applications. The feature extraction techniques discussed in this chapter mainly involve methods functioning through regression, filtering, clustering, transformation, and probabilistic theory.
