Fusion filter for hyperspectral target detection

Abstract

This paper presents work done using Schaum's fusion method, implemented in the CFAR (constant false alarm ratio) approach on a set of data in order to evaluate the success rate of this method with regard to common popular methods such as the GLRT, RX and a competing Projection Method created by the authors. The testing scheme involved the introduction of a free parameter in the form of a shading factor which affects similarly both the background and the target. Finally, the simulation results will be presented and discussed together with an analysis of the error regions from the fusion method implementation to outline the expected operation performance envelope for the fusion algorithm.