Share Email Print
cover

Proceedings Paper

Sparse hyperspectral unmixing combined L1/2 norm and reweighted total variation regularization
Author(s): Yan Li
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Sparse regression aims at estimating the fractional abundances of pure endmembers based on the assumption that each mixed pixel in the hyperspectral image can be expressed in the form of linear combinations of a number of known and pure endmembers. And total variation spatial regularization for sparse unmixing has been proposed with incorporating spatial information. In this paper, considering the desirable performance of reweighted minimization and owing to the L1/2 norm is an alternative regularizer which is much easier to solved than L0 regularizer and has better sparsity and robustness than L1 regularizer, a sparse regression combined L1/2 norm and reweighted total variation regularization has been utilized. Then the unconvex optimization problem is simply solved by the variable splitting and augmented Lagrangian algorithm. Our experimental results with simulated data sets and real hyperspectral data sets demonstrate that the proposed method is an effective and accurate spectral unmixing algorithm for hyperspectral regression.

Paper Details

Date Published: 21 July 2017
PDF: 9 pages
Proc. SPIE 10420, Ninth International Conference on Digital Image Processing (ICDIP 2017), 1042046 (21 July 2017); doi: 10.1117/12.2282140
Show Author Affiliations
Yan Li, Jiangxi Province Key Lab. of Precision Drive and Control (China)
Nanchang Institute of Technology (China)
Jiangxi Province Key Lab. of Water Information Cooperative Sensing and Intelligent Processing (China)


Published in SPIE Proceedings Vol. 10420:
Ninth International Conference on Digital Image Processing (ICDIP 2017)
Charles M. Falco; Xudong Jiang, Editor(s)

© SPIE. Terms of Use
Back to Top