A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images.
José Ignacio OrlandoElena ProkofyevaMatthew B. BlaschkoPublished in: IEEE Trans. Biomed. Eng. (2017)
Keyphrases
- random field model
- vessel segmentation
- fully connected
- fundus images
- blood vessels
- discriminatively trained
- retinal images
- random fields
- conditional random fields
- object detection
- generative model
- multiscale
- diabetic retinopathy
- optic disc
- discriminative learning
- low contrast
- discriminative training
- optical coherence tomography
- vessel enhancement
- scale free
- hidden markov models
- region of interest
- maximum entropy
- probabilistic model
- markov random field
- image segmentation
- pairwise
- image registration