Hyperspectral endmember spectra extraction based on constrained linear-quadratic matrix factorization using a projected gradient method.
Fatima Zohra BenhaloucheYannick DevilleMoussa Sofiane KarouiAbdelaziz OuamriPublished in: MLSP (2016)
Keyphrases
- matrix factorization
- gradient method
- hyperspectral images
- hyperspectral
- negative matrix factorization
- hyperspectral imagery
- hyperspectral data
- collaborative filtering
- multispectral
- infrared
- remote sensing
- low rank
- recommender systems
- missing data
- target detection
- convergence rate
- information content
- image data
- satellite images
- information extraction
- step size
- optimal control
- random projections
- multispectral images
- high resolution
- loss function
- principal component analysis
- multiresolution
- face recognition