Road Segmentation of Remotely-Sensed Images Using Deep Convolutional Neural Networks with Landscape Metrics and Conditional Random Fields.
Teerapong PanboonyuenKulsawasd JitkajornwanichSiam LawawirojwongPanu SrestasathiernPeerapon VateekulPublished in: Remote. Sens. (2017)
Keyphrases
- conditional random fields
- remotely sensed images
- road extraction
- convolutional neural networks
- image labeling
- segmentation method
- crf model
- change detection
- random fields
- graphical models
- remote sensing
- aerial images
- sequence labeling
- probabilistic model
- hidden markov models
- information extraction
- multispectral
- markov random field
- higher order
- supervised classification
- generative model
- structured prediction
- pairwise
- remote sensing images
- land cover
- segmentation algorithm
- image segmentation
- named entity recognition
- web page prediction
- region growing
- energy function
- satellite imagery
- image analysis
- superpixels
- object segmentation
- multispectral images
- level set
- reinforcement learning
- data sets
- object recognition
- multiscale
- bayesian networks
- data mining