A Research on the Combination Strategies of Multiple Features for Hyperspectral Remote Sensing Image Classification

Table 7

Classification results by different features for SL dataset, including OA, kappa index, and execution time.

SL		SPEC	GF	TF	GF + TF	TF + GF + SPEC
Direct combine	OA (%)	89.68	89.89	97.47	98.95	98.5
	Kappa	0.885	0.887	0.972	0.988	0.983
	Time (s)	10.9	3.88	5.98	6.63	9.7
LE framework	OA (%)	83.28	89.86	95.57	98.18	95.69
	Kappa	0.813	0.887	0.951	0.980	0.952
	Best	35	20	35	35	45
	Time (s)	11.21	5.42	7.82	7.29	9.743
Multiple kernel	OA (%)	89.68	90.01	96.83	97.74	97.34
	Kappa	0.885	0.899	0.964	0.974	0.970
	Time (s)	10.9	3.88	5.98	6.63	15.7