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

Table 6

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

IP		SPEC	GF	TF	GF + TF	TF + GF + SPEC
Direct combine	OA (%)	81.30	80.49	94.59	96.39	96.02
	Kappa	0.784	0.775	0.938	0.959	0.954
	Time (s)	3.3	0.99	1.46	1.77	4.18
LE framework	OA (%)	62.31	78.42	90.98	93.76	79.36
	Kappa	0.568	0.751	0.897	0.929	0.764
	Best	50	20	35	55	40
	Time (s)	4.23	2.44	2.73	3.38	4.06
Multiple kernel	OA (%)	81.30	85.89	93.72	94.36	92.88
	Kappa	0.784	0.838	0.928	0.935	0.918
	Time (s)	3.3	4.48	4.62	6.91	7.48