A Generalized Multiscale Bundle-Based Hyperspectral Sparse Unmixing Algorithm
This work addresses computational and noise issues in hyperspectral imaging for remote sensing applications, representing an incremental improvement over existing bundle-based methods.
The paper tackles the problem of hyperspectral sparse unmixing by addressing computational complexity and noise sensitivity in existing methods, proposing a generalized multiscale approach with group sparsity-inducing norms and a noise-robust method that ensures sparsity with reasonable cost, resulting in robust and consistent experimental outcomes compared to related methods.
In hyperspectral sparse unmixing, a successful approach employs spectral bundles to address the variability of the endmembers in the spatial domain. However, the regularization penalties usually employed aggregate substantial computational complexity, and the solutions are very noise-sensitive. We generalize a multiscale spatial regularization approach to solve the unmixing problem by incorporating group sparsity-inducing mixed norms. Then, we propose a noise-robust method that can take advantage of the bundle structure to deal with endmember variability while ensuring inter- and intra-class sparsity in abundance estimation with reasonable computational cost. We also present a general heuristic to select the \emph{most representative} abundance estimation over multiple runs of the unmixing process, yielding a solution that is robust and highly reproducible. Experiments illustrate the robustness and consistency of the results when compared to related methods.