Cardinality-constrained sparse spectral unmixing can be solved using Branch-and-Bound algorithms, provided that the number of reference endmembers and the cardinality constraint are reasonably small. However, focusing solely on the best solution may …
We focus on the exact resolution of sparse spectral unmixing problems, that is, the search for cardinality-limited linear least squares solutions under non-negativity and sum-to-one constraints. The originality of the proposed method - for which the Python code is provided - lies in its multisolution nature; we return the set of supports that yield the best solutions. The method is tested on synthetic data, with promising results.
We propose an algorithm that exactly solves the cardinality-constrained sparse spectral unmixing problem.
Implémentation de l'algorithme branch-and-bound pour le démélange spectral parcimonieux. [Matériel supplémentaire](https://gitlab.univ-nantes.fr/ls2n-sims/l0-sparse-unmix-eusipco-supplementary-materials). Version présentée à la conférence [EUSIPCO 2025](https://www.eusipco2025.org/) basée sur le prototype C++ [UNMIX](https://gitlab.com/mlatif/bbhs_ext_cpp)
09/2021 - Séminaire d'équipe SiMS.
Problèmes inverses, reconstruction statistique, optimisation et simulation Monte-Carlo.