A spectral clustering approach to underdetermined postnonlinear blind source separation of sparse sources.
Steven Van VaerenberghIgnacio SantamaríaPublished in: IEEE Trans. Neural Networks (2006)
Keyphrases
- blind source separation
- mixing matrix
- spectral clustering
- negative matrix factorization
- source separation
- eigenvalue decomposition
- blind separation
- source signals
- independent component analysis
- non stationary
- data clustering
- clustering method
- pairwise
- k means
- image segmentation
- sparse representation
- independent components
- clustering algorithm
- eigendecomposition
- graph laplacian
- spectral methods
- factor analysis
- laplacian matrix
- normalized cut
- speech signal
- denoising
- machine learning
- pattern recognition
- image processing