Creating Models Of Hyperspectral Classification Workflows Integrating Dimensionality Expansion For Multispectral Imagery

Dimensionality Expansion (DE) is a mathematical technique to increase multispectral data dimensionality in a nonlinear fashion so that standard hyperspectral linear methods can perform better at both pure-and mixed-pixel detection and classification. In this study, hyperspectral classification algorithms were applied on a WorldView-2 4-band image product. A visual modelling tool was used to build multiple classification workflows using subpixel target detection techniques that do not require knowledge of all endmembers in a scene, and to incorporate the DE into the classification process. The experiments have shown improved results for using the Adaptive Cosine/Coherence Estimator (ACE), Constrained Energy Minimization (CEM), and Matched Filter (MF) on the original WorldView-2 spectral bands plus the additional DE bands.