A Comparative Study of Likelihood Ratio Based Forensic Text Comparison in Procedures: Multivariate Kernel Density vs. Gaussian Mixture Model-Universal Background Model.
Shunichi IshiharaPublished in: ALTA (2013)
Keyphrases
- gaussian mixture model
- likelihood ratio
- background model
- background subtraction
- kernel density
- mixture model
- hypothesis testing
- gaussian model
- background modeling
- feature vectors
- motion detection
- maximum likelihood
- em algorithm
- mixture of gaussians
- mean shift
- foreground detection
- foreground objects
- gaussian distribution
- information retrieval
- probability density function
- gaussian mixture
- moving target
- density estimation
- pixel wise
- machine learning
- image segmentation
- moving objects
- hidden markov models
- video surveillance
- generative model
- expectation maximization
- pixel level
- statistical methods
- video sequences
- image sequences
- statistical tests