File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Regularization parameter selection in minimum volume hyperspectral unmixing

TitleRegularization parameter selection in minimum volume hyperspectral unmixing
Authors
Keywordsnonconvex optimization
hyperspectral images (HSIs)
spectral unmixing
Craig criterion
Issue Date2019
Citation
IEEE Transactions on Geoscience and Remote Sensing, 2019, v. 57, n. 12, p. 9858-9877 How to Cite?
AbstractLinear hyperspectral unmixing (HU) aims at factoring the observation matrix into an endmember matrix and an abundance matrix. Linear HU via variational minimum volume (MV) regularization has recently received considerable attention in the remote sensing and machine learning areas, mainly owing to its robustness against the absence of pure pixels. We put some popular linear HU formulations under a unifying framework, which involves a data-fitting term and an MV-based regularization term, and collectively solve it via a nonconvex optimization. As the former and the latter terms tend, respectively, to expand (reducing the data-fitting errors) and to shrink the simplex enclosing the measured spectra, it is critical to strike a balance between those two terms. To the best of our knowledge, the existing methods find such balance by tuning a regularization parameter manually, which has little value in unsupervised scenarios. In this paper, we aim at selecting the regularization parameter automatically by exploiting the fact that a too large parameter overshrinks the volume of the simplex defined by the endmembers, making many data points be left outside of the simplex and hence inducing a large data-fitting error, while a sufficiently small parameter yields a large simplex making data-fitting error very small. Roughly speaking, the transition point happens when the simplex still encloses the data cloud but there are data points on all its facets. These observations are systematically formulated to find the transition point that, in turn, yields a good parameter. The competitiveness of the proposed selection criterion is illustrated with simulated and real data.
Persistent Identifierhttp://hdl.handle.net/10722/298336
ISSN
2020 Impact Factor: 5.6
2020 SCImago Journal Rankings: 2.141
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZhuang, Lina-
dc.contributor.authorLin, Chia Hsiang-
dc.contributor.authorFigueiredo, Mario A.T.-
dc.contributor.authorBioucas-Dias, Jose M.-
dc.date.accessioned2021-04-08T03:08:11Z-
dc.date.available2021-04-08T03:08:11Z-
dc.date.issued2019-
dc.identifier.citationIEEE Transactions on Geoscience and Remote Sensing, 2019, v. 57, n. 12, p. 9858-9877-
dc.identifier.issn0196-2892-
dc.identifier.urihttp://hdl.handle.net/10722/298336-
dc.description.abstractLinear hyperspectral unmixing (HU) aims at factoring the observation matrix into an endmember matrix and an abundance matrix. Linear HU via variational minimum volume (MV) regularization has recently received considerable attention in the remote sensing and machine learning areas, mainly owing to its robustness against the absence of pure pixels. We put some popular linear HU formulations under a unifying framework, which involves a data-fitting term and an MV-based regularization term, and collectively solve it via a nonconvex optimization. As the former and the latter terms tend, respectively, to expand (reducing the data-fitting errors) and to shrink the simplex enclosing the measured spectra, it is critical to strike a balance between those two terms. To the best of our knowledge, the existing methods find such balance by tuning a regularization parameter manually, which has little value in unsupervised scenarios. In this paper, we aim at selecting the regularization parameter automatically by exploiting the fact that a too large parameter overshrinks the volume of the simplex defined by the endmembers, making many data points be left outside of the simplex and hence inducing a large data-fitting error, while a sufficiently small parameter yields a large simplex making data-fitting error very small. Roughly speaking, the transition point happens when the simplex still encloses the data cloud but there are data points on all its facets. These observations are systematically formulated to find the transition point that, in turn, yields a good parameter. The competitiveness of the proposed selection criterion is illustrated with simulated and real data.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Geoscience and Remote Sensing-
dc.subjectnonconvex optimization-
dc.subjecthyperspectral images (HSIs)-
dc.subjectspectral unmixing-
dc.subjectCraig criterion-
dc.titleRegularization parameter selection in minimum volume hyperspectral unmixing-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TGRS.2019.2929776-
dc.identifier.scopuseid_2-s2.0-85075667290-
dc.identifier.volume57-
dc.identifier.issue12-
dc.identifier.spage9858-
dc.identifier.epage9877-
dc.identifier.eissn1558-0644-
dc.identifier.isiWOS:000505701800028-
dc.identifier.issnl0196-2892-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats