Finding and visualizing relevant subspaces for clustering high-dimensional astronomical data using connected morphological operators.
Bilkis J. FerdosiHugo BuddelmeijerScott C. TragerMichael H. F. WilkinsonJos B. T. M. RoerdinkPublished in: IEEE VAST (2010)
Keyphrases
- morphological operators
- high dimensional
- high dimensional data
- mathematical morphology
- data points
- low dimensional
- high dimensionality
- flat zones
- morphological processing
- dimensionality reduction
- clustering method
- general theory
- spatially variant
- openings and closings
- clustering algorithm
- high dimensional data space
- high dimensional feature spaces
- lattice theory
- colour images
- parameter space
- subspace clustering
- nearest neighbor
- morphological filters
- k means
- structuring elements
- morphological operations
- partial differential equations
- translation invariant
- gray scale
- feature space
- multiscale
- bidimensional empirical mode decomposition
- image processing
- subspace clusters
- higher order
- wavelet transform