An iterative enhancement of higher order nonlinear mixture model for accurate hyperspectral unmixing

Abstract

In order to provide a careful description of the interactions among endmembers in hyperspectral images, a new method for adaptive design of mixture models for hyperspectral unmixing is introduced. Specifically, the proposed approach relies on exploiting geometrical features of hyperspectral signatures in terms of nonorthogonal projections onto the space induced by the endmembers' spectra. Then, an iterative process is deployed in order to understand the order of local nonlinearity that is displayed by each endmember over every pixel. Experimental results show that the proposed approach is actually able to retrieve thorough information on the nature of the nonlinear effects over the image while providing excellent performance in reconstructing the given dataset.