Scale-space approximated convolutional neural networks for retinal vessel segmentation.
Kyoung Jin NohSang Jun ParkSoochahn LeePublished in: Comput. Methods Programs Biomed. (2019)
Keyphrases
- scale space
- vessel segmentation
- convolutional neural networks
- retinal images
- blood vessels
- multiscale
- scale spaces
- gaussian scale space
- low contrast
- image analysis
- vessel enhancement
- scale selection
- image structure
- image features
- scale invariant
- multiple scales
- diffusion equation
- edge detector
- critical points
- region growing
- optic disc
- gaussian kernel
- region of interest
- vector valued
- scale space theory
- neural network
- higher order
- color images
- pattern recognition
- image segmentation