A Kernel-Based Feature Selection Method for SVM With RBF Kernel for Hyperspectral Image Classification.
Bor-Chen KuoHsin-Hua HoCheng-Hsuan LiChih-Cheng HungJin-Shiuh TaurPublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2014)
Keyphrases
- rbf kernel
- hyperspectral image classification
- radial basis function
- support vector
- kernel function
- support vector machine
- active learning
- hyperspectral images
- positive definite
- support vector machine svm
- hyperspectral
- feature space
- svm classifier
- reproducing kernel hilbert space
- cross validation
- basis functions
- neural network
- kernel methods
- remote sensing
- hyperspectral data
- loss function
- k nearest neighbor
- state space
- ls svm
- image data
- high dimensional
- hyperspectral imagery
- feature selection
- data mining