Ensemble Learning Based Multiple Kernel Principal Component Analysis for Dimensionality Reduction and Classification of Hyperspectral Imagery

Table 5

Overall accuracy, OA (%), and Kappa statistic, , of SVMs with different features. The best scores for each data set are highlighted in bold face font, and the second best is underlined.

Data	Kernel	Metric	KPCA							A-KPCA	M-KPCA
			Indian Pines	RBF	OA	83.54	88.69	90.52	92.69			57.80	52.41	19.47	94.09	96.03
	0.8006	0.8703			0.8918	0.9202	0.5036	0.4405	0.1267	0.9327	0.9459
KSC	Cauchy	OA	91.83	93.78	93.24	94.34	80.27	72.54	69.76	95.93	96.36
			0.9089	0.9307	0.9247	0.9409	0.7794	0.6918	0.6605	0.9547	0.9574