Unsupervised Nonlinear Hyperspectral Unmixing Based on Bilinear Mixture Models via Geometric Projection and Constrained Nonnegative Matrix Factorization.
Bin YangBin WangZongmin WuPublished in: Remote. Sens. (2018)
Keyphrases
- mixture model
- nonnegative matrix factorization
- hyperspectral
- unsupervised learning
- hyperspectral images
- remote sensing
- negative matrix factorization
- multispectral
- matrix factorization
- hyperspectral imagery
- em algorithm
- data representation
- hyperspectral data
- infrared
- probabilistic model
- image data
- spectral clustering
- semi supervised
- model selection
- expectation maximization
- language model
- supervised learning
- generative model
- maximum likelihood
- least squares
- principal component analysis
- information content
- dimensionality reduction
- image analysis
- document clustering
- satellite images
- original data
- machine learning
- feature selection
- objective function
- singular value decomposition
- training set
- object recognition
- data clustering
- image sequences
- image processing
- sparse representation
- computer vision
- information retrieval
- data sets