Weighting training images by maximizing distribution similarity for supervised segmentation across scanners.
Annegreet van OpbroekMeike W. VernooijMohammad Arfan IkramMarleen de BruijnePublished in: Medical Image Anal. (2015)
Keyphrases
- similarity measure
- image segmentation
- fully unsupervised
- multiscale
- segmentation method
- supervised learning
- image analysis
- semi supervised
- segmentation algorithm
- object segmentation
- learning algorithm
- similarity function
- distance measure
- medical images
- shape prior
- normalized cut
- region growing
- mixture of gaussian distributions
- level set
- active learning
- background subtraction
- euclidean distance
- random variables
- spatial distribution
- edge detection
- criterion function
- image processing