PolSAR Image Classification Using Hybrid Conditional Random Fields Model Based on Complex-Valued 3-D CNN.
Peng ZhangXiaofeng TanBeibei LiYinyin JiangWanying SongMing LiYan WuPublished in: IEEE Trans. Aerosp. Electron. Syst. (2021)
Keyphrases
- conditional random fields
- complex valued
- image classification
- real valued
- graphical models
- recurrent neural networks
- cellular neural networks
- linear equations
- hidden markov models
- maximum entropy
- higher order
- random fields
- probabilistic model
- sequence labeling
- crf model
- markov random field
- generative model
- feature extraction
- information extraction
- pairwise
- activation function
- multi valued
- multi label
- web page prediction
- neural network
- image features
- hidden layer
- feature selection
- random variables
- class labels
- knn
- data mining