Nonnegative Matrix Factorization with Data-Guided Constraints for Hyperspectral Unmixing.

Risheng Huang,Xiaorun Li,Liaoying Zhao
DOI: https://doi.org/10.3390/rs9101074
IF: 5
2017-01-01
Remote Sensing
Abstract:Hyperspectral unmixing aims to estimate a set of endmembers and corresponding abundances in pixels. Nonnegative matrix factorization (NMF) and its extensions with various constraints have been widely applied to hyperspectral unmixing. L-1/2 and L-2 regularizers can be added to NMF to enforce sparseness and evenness, respectively. In practice, a region in a hyperspectral image may possess different sparsity levels across locations. The problem remains as to how to impose constraints accordingly when the level of sparsity varies. We propose a novel nonnegative matrix factorization with data-guided constraints (DGC-NMF). The DGC-NMF imposes on the unknown abundance vector of each pixel with either an L-1/2 constraint or an L-2 constraint according to its estimated mixture level. Experiments on the synthetic data and real hyperspectral data validate the proposed algorithm.
What problem does this paper attempt to address?