Constrained feature selection for semisupervised color-texture image segmentation using spectral clustering.
Abderezak SalmiKamal HammoucheLudovic MacairePublished in: J. Electronic Imaging (2021)
Keyphrases
- spectral clustering
- texture image segmentation
- image segmentation
- feature selection
- active contours
- label propagation
- multiresolution
- fully automatic
- semi supervised
- clustering method
- pairwise
- data clustering
- graph partitioning
- k means
- similarity matrix
- evolutionary clustering
- computer vision
- constrained spectral clustering
- model selection
- feature extraction
- support vector
- feature set
- unsupervised learning
- affinity matrix
- graph laplacian
- training data
- normalized cut
- text categorization
- level set
- eigendecomposition
- spectral methods
- graph cuts
- support vector machine
- feature space
- semi automatic
- pairwise constraints
- expectation maximization
- clustering algorithm
- alternative clusterings
- supervised learning
- dimension reduction
- labeled data
- image processing