Total Variation Spatial Regularization for Sparse Hyperspectral Unmixing

By Marian-Daniel Iordache, José M. Bioucas-Dias, Antonio J. Plaza. IEEE Transactions on Geoscience and Remote Sensing, 2012

728 citations recorded by the source corpus.

Subfields: Computational Mechanics, Media Technology

View Paper: https://www.semanticscholar.org/paper/2b07d0bfcbc64b8cefa2d5d7b84a3dd13b6d7b51

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Media Technology

Citations in the corpus

Advances in Hyperspectral Image and Signal Processing: A Comprehensive Overview of the State of the Art (2017, methodology): “To incorporate spatial information into the spectral unmixing formulation, a TV regularizer has been developed to enforce spatial homogeneity by including this term in the original objective function [162].”
Collaborative Sparse Regression for Hyperspectral Unmixing (2014, methodology): “The TV regularizer was also used recently in hyperspectral unmixing (see [41]) and, despite the relatively high computational complexity of the model, it brings important improvements in the quality of the final abundance maps.”
Nonlinear Unmixing of Hyperspectral Images: Models and Algorithms (2013, methodology): “ribed hereafter. Regularization can also be used for integrating spatial information into the hyperspectral unmixing problem. In [38], the authors have proposed using ‘ 1-type regularizer of the form [41] to promote piecewise-constant transitions in the fractional abundances that parameterize April ”
A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote Sensing (2014, background): “Therefore, one can exploit such spatial/spectral contextual information for improving SR performance by applying piecewise smooth regularization, such as total variations (TVs) [57], on top of an SR formulation, e.”
Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches (2012, positive): “The limitation imposed to the sparse regression methods by the usual high correlation of the hyperspectral signatures is mitigated in [213], [214] by adding the Total Variation [211] regularization term, applied to the individual bands, to CSR problem (21).”
The Why and How of Nonnegative Matrix Factorization (2014, background): “…or spatial coherence (neighboring pixels are more likely to contain the same materials) which are usually tackled with TV-like regularizations (that is, ℓ 1 norm of the diﬀerence between neighboring values to preserve the edges in the image); see, e.g., [68, 67], and the references therein.”
Multitask Diffusion Adaptation Over Networks (2013, methodology): “This algorithm clearly contrasts with existing batch approaches based on FISTA [70] and ADMM [71], which cannot easily address large problems (84).”
Total-Variation-Regularized Low-Rank Matrix Factorization for Hyperspectral Image Restoration (2016, methodology): “3) The augmented Lagrange multiplier (ALM) [36]–[38] method is utilized and extended to solve the proposed model.”