Feature Normalization via Expectation Maximization and Unsupervised Nonparametric Classification For M-FISH Chromosome Images.
Hyohoon ChoiAlan C. BovikKenneth R. CastlemanPublished in: IEEE Trans. Medical Imaging (2008)
Keyphrases
- image classification
- image features
- preprocessing stage
- feature vectors
- expectation maximization
- dissimilarity representation
- image database
- image data
- three dimensional
- feature representation
- discriminative features
- input image
- unsupervised learning
- em algorithm
- image analysis
- classification accuracy
- fish species
- image retrieval
- pattern recognition
- supervised learning
- data driven
- image registration
- feature extraction
- thematic mapper
- support vector machine
- mixture model
- test images
- image set
- supervised classification
- feature set
- image processing
- pigmented skin lesions
- preprocessing steps
- highly discriminative
- feature maps
- hyperspectral images
- object recognition
- density estimation
- preprocessing
- maximum likelihood
- text classification
- machine learning
- gabor filters
- support vector
- textural features
- similarity measure
- feature space
- edge detection