2d mixture-of-Gaussians one-class classification (RG plane)
Visualize distributions
EM experiment #1: object 1 (Kellogs car)
Extract 2d data
Choose number of Gaussians in mixture model
Assumes full covariance matrices
Compute EM algorithm
Save solution
Final model
One-class image classification for different thresholds (0.5 to 5.0)
Error plot for different threshold (0.5 to 5.0)
Created by Mathematica (October 9, 2003)