Sparse Nonnegative Matrix Factorization for Hyperspectral Unmixing Based on Endmember Independence and Spatial Weighted Abundance.
Jingyan ZhangXiangrong ZhangLicheng JiaoPublished in: Remote. Sens. (2021)
Keyphrases
- nonnegative matrix factorization
- hyperspectral
- hyperspectral images
- spectral signatures
- hyperspectral imagery
- hyperspectral data
- sparsity constraints
- remote sensing
- infrared
- negative matrix factorization
- random projections
- multispectral
- data representation
- matrix factorization
- target detection
- image data
- principal component analysis
- least squares
- document clustering
- original data
- information content
- spectral clustering
- spatial resolution
- satellite images
- sparse representation
- data sets
- image analysis
- similarity measure
- collaborative filtering
- pairwise
- objective function
- computer vision