Noise Identification and Estimation of its Statistical Parameters by Using Unsupervised Variational Classification.
Benoît VozelKacem ChehdiLuc KlaineVladimir V. LukinSergey K. AbramovPublished in: ICASSP (2) (2006)
Keyphrases
- unsupervised learning
- parameter estimation
- supervised learning
- data driven
- decision trees
- classification accuracy
- estimation process
- unsupervised clustering
- automatic classification
- classification method
- maximum likelihood criterion
- unsupervised classification
- pattern recognition
- image classification
- class labels
- parameter identification
- kernel density estimators
- classification algorithm
- maximum entropy modeling
- support vector machine svm
- feature extraction
- input data
- machine learning
- partially supervised
- feature space
- feature vectors
- support vector machine
- maximum likelihood estimation
- parametric models
- semi supervised
- estimation algorithm
- density estimation
- pattern classification
- text classification
- cross validation
- noise level
- signal to noise ratio
- multiscale
- image segmentation
- feature selection