Spectral Super-Resolution of Hyperspectral Imagery Using Re-Weighted L1 Spatial Filtering (bibtex)
by and
Abstract:
Sparsity-based models have enabled significant advances in many image processing tasks. Hyperspectral imagery (HSI) in particular has benefited from these approaches due to the significant low-dimensional structure in both spatial and spectral dimensions. Specifically, previous work has shown that sparsity models can be used for spectral super-resolution, where spectral signatures with HSI-level resolution are recovered from measurements with multispectral-level resolution (i.e., an order of magnitude fewer spectral bands). In this paper we expand on those results by introducing a new inference approach known as re-weighted l1 spatial filtering (RWL1-SF). RWL1-SF incorporates a more sophisticated signal model that allows for variations in the SNR at each pixel as well as spatial dependencies between neighboring pixels. The results demonstrate that the proposed approach leverages signal structure beyond simple sparsity to achieve significant improvements in spectral super-resolution.
Reference:
Spectral Super-Resolution of Hyperspectral Imagery Using Re-Weighted L1 Spatial FilteringA.S. Charles and C.J. Rozell. IEEE Geoscience and Remote Sensing Letters, 11(3), pp. 602–606, March 2014.
Bibtex Entry:
@Article{charles.12i,
  author = 	 {Charles, A.S. and Rozell, C.J.},
  title = 	 {Spectral Super-Resolution of Hyperspectral Imagery Using Re-Weighted {L1} Spatial Filtering},
  abstract =     {
Sparsity-based models have enabled significant advances in many image processing tasks. Hyperspectral imagery (HSI) in particular has benefited from these approaches due to the significant low-dimensional structure in both spatial and spectral dimensions. Specifically, previous work has shown that sparsity models can be used for spectral super-resolution, where spectral signatures with HSI-level resolution are recovered from measurements with multispectral-level resolution (i.e., an order of magnitude fewer spectral bands).  In this paper we expand on those results by introducing a new inference approach known as re-weighted l1 spatial filtering (RWL1-SF).  RWL1-SF incorporates a more sophisticated signal model that allows for variations in the SNR at each pixel as well as spatial dependencies between neighboring pixels.  The results demonstrate that the proposed approach leverages signal structure beyond simple sparsity to achieve significant improvements in spectral super-resolution.},
year = 2014,
journal = {IEEE Geoscience and Remote Sensing Letters},
url = {http://siplab.gatech.edu/pubs/charlesGRSL_apr_2013.pdf},
volume = 11,
number = 3,
pages = {602--606},
month = mar
}
Powered by bibtexbrowser