Hyperspectral Unmixing Using Sparsity-Constrained Deep Nonnegative Matrix Factorization With Total Variation
Hyperspectral Unmixing Using Sparsity-Constrained Deep Nonnegative Matrix Factorization With Total Variation is a scholarly work, published in 2018 in ''IEEE Transactions on Geoscience and Remote Sensing''. The main subjects of the publication include Abundance estimation, non-negative matrix factorization, end member, artificial intelligence, image fusion, computer science, remote sensing, algorithm, constraint, matrix decomposition, pixel, mathematics, pattern recognition, and hyperspectral imaging. To alleviate such limitation, in this paper, authors propose a new sparsity-constrained deep NMF with total variation (SDNMF-TV) technique for hyperspectral unmixing.