Dimensionality reduction, classification, and spectral mixture analysis using non-negative underapproximation
Authors:
- 1 February 2011
- journal article
- Published by SPIE-Intl Soc Optical Eng in Optical Engineering
- Vol. 50 (2) , 027001-027001-16
Abstract
Non-negative matrix factorization (NMF) and its variants have recently been successfully used as dimensionality reduction techniques for identification of the materials present in hyperspectral images. We study a recently introduced variant of NMF called non-negative matrix underapproximation (NMU): it is based on the introduction of underapproximation constraints, which enables one to extract features in a recursive way, such as principal component analysis, but preserving non-negativity. We explain why these additional constraints make NMU particularly well suited to achieve a parts-based and sparse representation of the data, enabling it to recover the constitutive elements in hyperspectral images. Both ℓ2-norm and ℓ1-norm–based minimization of the energy functional are considered. We experimentally show the efficiency of this new strategy on hyperspectral images associated with space object material identification, and on HYDICE and related remote sensing images.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- On the Complexity of Nonnegative Matrix FactorizationSIAM Journal on Optimization, 2010
- Tensor methods for hyperspectral data analysis: a space object material identification studyJournal of the Optical Society of America A, 2008
- Theorems on Positive Data: On the Uniqueness of NMFComputational Intelligence and Neuroscience, 2008
- Algorithms and applications for approximate nonnegative matrix factorizationComputational Statistics & Data Analysis, 2007
- Non-negative Matrix Factorization with Orthogonality Constraints and its Application to Raman SpectroscopyJournal of Signal Processing Systems, 2007
- Endmember Extraction From Highly Mixed Data Using Minimum Volume Constrained Nonnegative Matrix FactorizationIEEE Transactions on Geoscience and Remote Sensing, 2007
- Orthogonal nonnegative matrix t-factorizations for clustering2006
- Nonnegative matrix factorization for spectral data analysisLinear Algebra and its Applications, 2006
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- Minimum-volume transforms for remotely sensed dataIEEE Transactions on Geoscience and Remote Sensing, 1994