Hyperspectral Image Classification Based on a Convolutional Neural Network and Discontinuity Preserving Relaxation.
Qishuo GaoSamsung LimPublished in: IGARSS (2018)
Keyphrases
- convolutional neural network
- discontinuity preserving
- hyperspectral image classification
- surface reconstruction
- active learning
- optical flow estimation
- hyperspectral
- face detection
- graph cuts
- hyperspectral images
- motion field
- markov random field
- nonlinear diffusion
- map estimation
- smoothness constraint
- energy minimization
- maximum a posteriori
- neural network
- stereo correspondence
- optical flow
- depth discontinuities
- energy function
- maximum likelihood
- phase unwrapping
- learning algorithm
- moving objects
- range data
- motion compensation
- mutual information
- denoising
- video sequences