Sparse Tensor-Based Dimensionality Reduction for Hyperspectral Spectral-Spatial Discriminant Feature Extraction.
Zhi LiuBo TangXiaofu HeQingchen QiuHongjun WangPublished in: IEEE Geosci. Remote. Sens. Lett. (2017)
Keyphrases
- feature extraction
- hyperspectral
- spectral signatures
- dimensionality reduction
- frequency domain
- hyperspectral data
- random projections
- principal component analysis
- subspace learning
- remote sensing
- hyperspectral images
- spatial frequency
- hyperspectral imagery
- discriminant analysis
- low dimensional
- multispectral
- remote sensing data
- sparse representation
- infrared
- target detection
- spectral imagery
- high dimensional
- discriminant information
- hyperspectral remote sensing
- spectral bands
- dimension reduction
- image data
- feature space
- pattern recognition
- satellite images
- locality preserving projections
- spectral resolution
- image processing
- feature vectors
- manifold learning
- spectral data
- high dimensional data
- linear discriminant analysis
- face recognition
- hyperspectral imaging
- spatial information
- spatial resolution
- image classification
- feature selection
- hyperspectral image classification
- principal components
- high dimensionality
- dimensionality reduction methods
- pixel classification
- singular value decomposition
- wavelet transform
- multi band
- feature set
- remote sensing imagery
- data points
- texture features
- principle component analysis
- band selection
- information content
- high quality
- reflectance spectra
- kernel pca
- change detection