Shearlet-Based Structure-Aware Filtering for Hyperspectral and LiDAR Data Classification

Table 8

Classification performance using Raw-H, Raw, 3D-CNN, miniGCN, SAE-LR, NMFL, EPCA, GGF, EMAP, OTVCA, and ShearSAF for the MUUFL Gulfport dataset with five labeled samples per class as the training set.


Class	Raw-H	Raw	3D-CNN	MiniGCN	SAE-LR	NMFL	EPCA	GGF	EMAP	OTVCA	ShearSAF

C1	67.06	70.45	83.15	79.98	68.09	69.77	63.27	61.10	70.13	75.05	83.52
C2	70.85	71.44	68.84	90.58	58.11	67.42	70.54	72.06	73.26	53.32	72.54
C3	42.47	42.95	37.43	38.60	30.16	50.18	36.47	52.52	40.68	32.49	58.26
C4	47.33	47.38	61.81	59.96	61.73	68.54	46.71	49.72	61.84	63.78	65.74
C5	69.63	72.68	66.42	81.13	57.12	73.75	76.04	79.67	80.58	54.02	82.70
C6	84.24	86.15	61.03	94.41	65.21	98.40	97.99	96.78	96.75	99.87	87.55
C7	62.38	68.86	71.62	76.19	55.73	72.29	82.60	76.64	77.69	74.99	89.25
C8	32.23	39.17	74.54	78.41	79.28	59.27	73.59	52.81	70.90	66.80	64.08
C9	42.46	43.83	38.06	40.08	29.70	45.47	39.57	45.62	39.87	34.40	48.61
C10	61.07	61.64	49.21	76.40	40.82	84.78	53.33	61.45	59.73	69.40	61.72
C11	90.37	90.65	61.23	72.30	42.86	74.31	93.09	88.90	94.33	96.49	89.96
OA	59.23	62.32	70.28	68.69	60.40	66.11	63.25	62.53	68.05	64.60	72.30
Kappa	0.50	0.53	0.62	0.62	0.51	0.58	0.55	0.54	0.60	0.56	0.65