Publisher's Synopsis
The purpose of this research was to expand the current catalog of hyperspectral imagery (HSI) anomaly detector algorithms by proving the effectiveness of a new detector dubbed Iterative RX. Specifically, the Iterative RX detector is measured against the established HSI "benchmark" detector known as RX. Principal component analysis was employed as a data reduction technique. Design of experiments was used to ascertain the significance of three factors including algorithm used. Detector performance was measured both by the standard operating characteristic curve and by the novel clump operating characteristic curve. Matlab code for the Iterative RX algorithm, Baseline RX algorithm, and a clump grading function are provided. Furthermore, recommended algorithm settings and procedures for the application of the Iterative RX detector are also provided.
