A NEW EXTENDED LINEAR MIXING MODEL TO ADDRESS SPECTRAL VARIABILITY

Miguel A. Veganzones, Lucas Drumetz, Guillaume Tochon, Mauro Dalla Mura, Antonio Plaza, José M. Bioucas‐Dias, Jón Atli Benediktsson

Abstract

Spectral variability is a phenomenon due, to a grand extend, to varia-tions in the illumination and atmospheric conditions within a hyper-spectral image, causing the spectral signature of a material to vary within a image. Data spectral fluctuation due to spectral variabil-ity compromises the linear mixing model (LMM) sum-to-one con-straint, and is an important source of error in hyperspectral image analysis. Recently, spectral variability has raised more attention and some techniques have been proposed to address this issue, i.e. spec-tral bundles. Here, we propose the definition of an extended LMM (ELMM) to model spectral variability and we show that the use of spectral bundles models the ELMM implicitly. We also show that the constrained least squares (CLS) is an explicit modelling of the ELMM when the spectral variability is due to scaling effects. We give experimental validation that spectral bundles (and sparsity) and CLS are complementary techniques addressing spectral variability. We finally discuss on future research avenues to fully exploit the proposed ELMM. Index Terms — Spectral unmixing, extended linear mixing model, spectral bundles, sparsity, CLS. 1.
BibTeX
@inproceedings{veganzones2014new,
  author = {Miguel A. Veganzones and Lucas Drumetz and Guillaume Tochon and Mauro Dalla Mura and Antonio Plaza and José M. Bioucas‐Dias and Jón Atli Benediktsson},
  title = {{A NEW EXTENDED LINEAR MIXING MODEL TO ADDRESS SPECTRAL VARIABILITY}},
  booktitle = {WHISPERS},
  year = {2014},
  url = {https://openreview.net/forum?id=dDoPI3Krud},
}

Loading related content...

Figure