Sparse Unmixing using an approximate L0 Regularization

DOI

10.2991/icismme-15.2015.189How to use a DOI?

Keywords

Sparse unmixing; approximate sparsity; the linear mixture model; and approximate sparsity regularizer

Abstract

Recently, sparse unmixing focuses on finding an optimal subset of spectral signatures in a large spectral spetral library. In most previous work concerned with the sparse unmixing, the linear mixture model has been widely used to determine and quantify the abundance of materials in mixed piexels[1]. In this paper, we propose a new sparse unmxing method based on an approximate sparsity regularization model[2]. The approximate sparsity regularizer is much easier to solve than the L0 regularizer and has stronger sparsity than the L1 regularizer. What’s more, a variable splitting and augmented Lagrangian methods introduced in to solve the proposed problem. Our numerical results on sparse unmixing illustration the efficiency of approximate sparsity a under the SUnSAL algorithm framework, compared to the L1 norm.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yang Guo
AU  - Tai Gao
AU  - Chengzhi Deng
AU  - Shengqian Wang
AU  - JianPing Xiao
PY  - 2015/07
DA  - 2015/07
TI  - Sparse Unmixing using an approximate L0 Regularization
BT  - Proceedings of the First International Conference on Information Sciences, Machinery, Materials and Energy
PB  - Atlantis Press
SP  - 899
EP  - 903
SN  - 1951-6851
UR  - https://doi.org/10.2991/icismme-15.2015.189
DO  - 10.2991/icismme-15.2015.189
ID  - Guo2015/07
ER  -