EURASIP Journal on Advances in Signal Processing

Table 4 Recognition rate (percent) for biopsy image data from different one-class classifier ensembles

From: One-class kernel subspace ensemble for medical image classification

Image class	PCA	MoG	KMeans	SVDD	Parzen	KPCA
Normal	85.17	82.12	80.12	85.56	84.54	92.70
In situ	87.33	84.67	83.46	87.22	81.26	93.78
Invasive	82.56	81.88	79.65	84.67	83.23	90.35

The kernel widths for KPCA and SVDD were set to σ = 4. The number of principal components for KPCA and PCA were set to n = 40.

Back to article page