Hyperspectral Target Detection Using Gram-Schmidt Orthogonal Projections
5 Pages Posted: 8 Aug 2019
Date Written: August 7, 2019
Abstract
The automatic detection of targets in the hyperspectral images is highly used in defense and security applications. Hyperspectral images are rich in information due to its high spectral and spatial resolution. In this paper, we use Automatic Target Generation Process (ATGP) to extract the endmembers from the hyperspectral image. We compared the performance of Gram-Schmidt orthogonal vector projection (GSOVP) with the classical Orthogonal Subspace Projection (OSP). In our experiments, we used HYDICE and ROSIS urban datasets for evaluating the performance of the optimized algorithm.
Keywords: hyperspectral imaging, orthogonal projection, automatic target detection, Gram-Schmidt method
Suggested Citation: Suggested Citation