Robust RX anomaly detector without covariance matrix estimation.
Santiago Velasco-ForeroJesús AnguloPublished in: WHISPERS (2012)
Keyphrases
- covariance matrix
- estimation error
- covariance matrices
- principal component analysis
- geometrical interpretation
- sample size
- mahalanobis distance
- eigenvalues and eigenvectors
- positive definite
- multivariate normal
- pseudo inverse
- gaussian mixture
- multivariate gaussian
- anomaly detection
- class conditional densities
- computer vision
- parameter estimation
- feature vectors
- objective function