期刊
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
卷 12, 期 11, 页码 2311-2315出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LGRS.2015.2474744
关键词
Abundance estimation; hyperspectral image; similarity-weighting; sparse unmixing; l(p)-regularization
类别
资金
- 973 Program [2011CB707104]
- National Science Foundation of China [61372147, 61501188]
- Science Foundation of Shanghai [15ZR1410200]
In this letter, we propose a novel sparse unmixing model combined with two effective regularization terms: one is a similarity-weighting constraint, and the other is the l(p) (0 < p < 1) norm sparse regularization. The former utilizes the spatial structural correlation, which is presented in the hyperspectral data, to guide the abundance estimation. When compared with the existing graph Laplacian regularization, our similarity-weighting constraint avoids large matrix inversion, and thus, it can be efficiently solved. As for the l(p)-norm, it has numerical advantages over the convex l(1)-norm and better approximates the l(0)-norm theoretically. Moreover, the l(p)-norm regularizer can simultaneously promote sparsity and enforce the abundance sum-to-one constraint. Therefore, this term yields more desirable results in practice. Experimental results on both simulated and real data demonstrate the effectiveness of the proposed model.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据