Multiscale Superpixel-Level Subspace-Based Support Vector Machines for Hyperspectral Image Classification.
Haoyang YuLianru GaoWenzhi LiaoBing ZhangAleksandra PizuricaWilfried PhilipsPublished in: IEEE Geosci. Remote. Sens. Lett. (2017)
Keyphrases
- multiscale
- support vector
- hyperspectral image classification
- learning machines
- superpixels
- image processing
- image segmentation
- large margin classifiers
- active learning
- feature extraction
- wavelet transform
- image representation
- principal component analysis
- hyperspectral images
- low dimensional
- support vector machine
- kernel function
- wavelet coefficients
- kernel methods
- image fusion
- data sets
- image analysis
- reinforcement learning
- learning algorithm