An active learning method based on Markov random fields for hyperspectral images classification.
Shujin SunPing ZhongHuaitie XiaoFang LiuRunsheng WangPublished in: WHISPERS (2015)
Keyphrases
- markov random field
- hyperspectral images
- pairwise
- active learning
- parameter estimation
- energy function
- map estimation
- maximum a posteriori
- bayesian networks
- higher order
- edge detection
- image segmentation
- graph cuts
- segmentation method
- maximum likelihood
- feature selection
- belief propagation
- image restoration
- model selection
- image processing
- wavelet transform
- object recognition
- similarity measure