Dc. Heinz et Ci. Chang, Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery, IEEE GEOSCI, 39(3), 2001, pp. 529-545
Citations number
39
Categorie Soggetti
Eletrical & Eletronics Engineeing
Journal title
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
Linear spectral mixture analysis (LSMA) is a widely used technique in remot
e sensing to estimate abundance fractions of materials present in an image
pixel. In order for an LSMA-based estimator to produce accurate amounts of
material abundance, it generally requires two constraints imposed on the li
near mixture model used in LSMA, which are the abundance sum-to-one constra
int and the abundance nonnegativity constraint. The first constraint requir
es the sum of the abundance fractions of materials present in an image pixe
l to be one and the second imposes a constraint that these abundance fracti
ons be nonnegative. While the first constraint is easy to deal with, the se
cond constraint is difficult to implement since it results in a set of ineq
ualities and can only be solved by numerical methods. Consequently, most LS
MA-based methods are unconstrained and produce solutions that do not necess
arily reflect the true abundance fractions of materials. In this case, they
can only be used for the purposes of material detection, discrimination, a
nd classification, but not for material quantification. In this paper, we p
resent a fully constrained least squares (FCLS) linear spectral mixture ana
lysis method for material quantification. Since no closed form can be deriv
ed for this method, an efficient algorithm is developed to yield optimal so
lutions. In order to further apply the designed algorithm to unknown image
scenes, an unsupervised least squares error (LSE)-based method is also prop
osed to extend the FCLS method in an unsupervised manner. A series of compu
ter simulations and real hyperspectral data experiments were conducted to d
emonstrate the performance of the proposed FCLS LSMA approach in material q
uantification.