Adversarial U-net with spectral normalization for histopathology image segmentation using synthetic data.
Faisal MahmoodRichard J. ChenDaniel BordersGregory N. McKayKevan J. SalimianAlexander S. BarasNicholas J. DurrPublished in: Medical Imaging: Digital Pathology (2019)
Keyphrases
- synthetic data
- image segmentation
- normalized cut
- data sets
- graph cuts
- real image data
- mri data
- deformable models
- image processing
- method for image segmentation
- real world
- gray level
- markov random field
- preprocessing
- active contours
- contour detection
- segmentation method
- spectral clustering
- boundary detection
- level set method
- hyperspectral images
- spectral features
- multispectral images
- hyperspectral
- computer vision
- energy functional
- graph partitioning
- shape prior
- region growing
- medical imaging
- unsupervised image segmentation
- normalization method
- image analysis
- level set
- hyperspectral imagery
- region segmentation
- sea surface
- multiscale
- watershed transformation
- unsupervised segmentation
- spectral analysis
- expectation maximization
- segmented images
- energy function
- texture segmentation
- fuzzy c means