Conditional Random Fields for Multitemporal and Multiscale Classification of Optical Satellite Imagery.
Thorsten HobergFranz RottensteinerRaul Queiroz FeitosaChristian HeipkePublished in: IEEE Trans. Geosci. Remote. Sens. (2015)
Keyphrases
- conditional random fields
- satellite imagery
- change detection
- multiscale
- remote sensing
- electro optical
- satellite images
- random fields
- probabilistic model
- information extraction
- structured output
- sequence labeling
- graphical models
- hidden markov models
- named entity recognition
- feature extraction
- markov random field
- pattern recognition
- higher order
- superpixels
- generative model
- pairwise
- remotely sensed images
- crf model
- machine learning
- model selection
- image classification
- image registration
- support vector
- web page prediction
- multispectral
- data sets
- image processing
- semi markov
- decision trees
- training set
- image analysis
- land cover
- urban areas
- supervised classification
- remote sensing images
- support vector machine
- supervised learning