Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images

Table 6

Comparison of BDT-SMO classification approaches under different kernels (zone 1).

Kernels	Overall accuracy (%)	Average accuracy (%)	Kappa coefficient	Training time (s)	Test time (s)	Standard deviation	Optimized parameters
Linear	79.34	79.39	0.7752	35.24	18.49	1.88
Polynomial	80.05	80.41	0.7841	39.79	20.63	1.79	,
Sigmoid	82.12	82.55	0.8024	38.15	22.94	1.52	,
SRBF	83.46	83.66	0.8195	30.57	13.67	0.65	,
CRBF	85.43	85.67	0.8411	33.27	12.44	0.63	,; ,