Segmentation of angiodysplasia lesions in WCE images using a MAP approach with Markov Random Fields.
Pedro M. VieiraBruno GonçalvesCarla R. GoncalvesCarlos S. LimaPublished in: EMBC (2016)
Keyphrases
- markov random field
- maximum a posteriori
- mrf model
- map estimation
- maximum a posteriori probability
- textured images
- image segmentation
- segmentation algorithm
- markov random field model
- image analysis
- random fields
- graph cuts
- energy function
- segmentation method
- mrf models
- low level vision
- unsupervised segmentation
- prior models
- gibbs distribution
- energy minimization
- parameter estimation
- test images
- edge detection
- image restoration
- prior model
- label field
- maximum a posterior
- higher order
- belief propagation
- input image
- skin lesion
- texture segmentation
- image segmentation algorithms
- pairwise
- multiscale
- smoothness prior
- object recognition
- image understanding
- scene labeling
- iterated conditional modes
- region growing
- computer aided
- image features
- color images
- piecewise constant functions
- medical images
- level set
- bayesian networks
- computer vision
- discriminative random fields
- gauss markov random fields
- capsule endoscopy
- colour images
- map inference
- image edges
- outdoor scenes
- shape prior
- maximum entropy
- natural images
- texture features
- image reconstruction
- conditional random fields
- gray level