Novelty detection and segmentation based on Gaussian mixture models: A case study in 3D robotic laser mapping.
Paulo Drews Jr.Pedro NúñezRui P. RochaMario Fernando Montenegro CamposJorge DiasPublished in: Robotics Auton. Syst. (2013)
Keyphrases
- novelty detection
- gaussian mixture model
- mixture model
- text filtering
- maximum likelihood criterion
- speaker recognition
- image segmentation
- anomaly detection
- bayesian information criterion
- concept drift
- maximum likelihood
- covariance matrices
- em algorithm
- probability density
- feature vectors
- speaker identification
- feature space
- expectation maximization
- sentence level
- probability density function
- image analysis
- density estimation
- finite mixtures
- machine learning