scholarly journals Spectral unmixing using nonnegative matrix factorization with smoothed L0 norm constraint

2009 ◽  
Author(s):  
Zuyuan Yang ◽  
Xi Chen ◽  
Guoxu Zhou ◽  
Shengli Xie
Keyword(s):  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Spectral Unmixing ◽  
Norm Constraint

Nonnegative Matrix Factorization with Fixed L2-Norm Constraint

2019 ◽  
Vol 38 (7) ◽  
pp. 3211-3226 ◽  
Author(s):  
Zuyuan Yang ◽  
Yifei Hu ◽  
Naiyao Liang ◽  
Jun Lv
Keyword(s):  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
L2 Norm ◽  
Norm Constraint

Blind spectral unmixing in terahertz domain using nonnegative matrix factorization

2015 ◽  
Author(s):  
Xian Li ◽  
Ping J. Huang ◽  
Ye H. Ma ◽  
Di B. Hou ◽  
Guangxin Zhang
Keyword(s):  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Spectral Unmixing

scholarly journals Regularized nonnegative matrix factorization: Geometrical interpretation and application to spectral unmixing

2014 ◽  
Vol 24 (2) ◽  
pp. 233-247 ◽  
Author(s):  
Rafał Zdunek
Keyword(s):  
Matrix Factorization ◽  
Numerical Optimization ◽  
Nonnegative Matrix Factorization ◽  
Regularization Parameter ◽  
Nonnegative Matrix ◽  
Spectral Unmixing ◽  
Geometrical Analysis ◽  
Observation Matrix ◽  
Mixing Matrix ◽  
Low Performance

Abstract Nonnegative Matrix Factorization (NMF) is an important tool in data spectral analysis. However, when a mixing matrix or sources are not sufficiently sparse, NMF of an observation matrix is not unique. Many numerical optimization algorithms, which assure fast convergence for specific problems, may easily get stuck into unfavorable local minima of an objective function, resulting in very low performance. In this paper, we discuss the Tikhonov regularized version of the Fast Combinatorial NonNegative Least Squares (FC-NNLS) algorithm (proposed by Benthem and Keenan in 2004), where the regularization parameter starts from a large value and decreases gradually with iterations. A geometrical analysis and justification of this approach are presented. The numerical experiments, carried out for various benchmarks of spectral signals, demonstrate that this kind of regularization, when applied to the FC-NNLS algorithm, is essential to obtain good performance.

Spectral unmixing with nonnegative matrix factorization

2006 ◽  
Author(s):  
Mario Parente ◽  
Argyrios Zymnis ◽  
Joëlle Skaf ◽  
Janice Bishop
Keyword(s):  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Spectral Unmixing
Sign in / Sign up

Export Citation Format

Share Document