A Slightly Supervised Approach for Positive/Negative Classification of Fluorescence Intensity in HEp-2 Images.
Giulio IannelloLeonardo OnofriPaolo SodaPublished in: ICIAP (2) (2013)
Keyphrases
- image analysis
- image classification
- positive and negative
- input image
- image features
- image database
- intensity values
- supervised classification
- image registration
- pattern recognition
- unsupervised learning
- classification accuracy
- image data
- machine learning
- supervised learning
- decision trees
- decision rules
- preprocessing stage
- object recognition
- confocal microscopy
- thematic mapper
- image gradient
- feature selection
- intensity difference
- pixel intensities
- skin lesion
- intensity levels
- image annotation
- image set
- feature vectors
- three dimensional
- intensity images
- test images
- image regions
- single image
- intensity variations
- segmentation algorithm
- active learning
- unsupervised methods
- microscope images
- optical properties
- feature space
- support vector