Hyperspectral Image Classification with Optimized Compressed Synergic Deep Convolution Neural Network with Aquila Optimization

Table 1

HSI categorization for Indiana Pines (IP) dataset.

Methods	RNN	DCNN	SDCNN	CSDCNN	CSDCNN-AO
OA	46.33 ± 0.45	48.73 ± 0.89	89.36 ± 1.13	89.57 ± 0.86	93.44 ± 1.08
AA	36.20 ± 1.06	49.60 ± 3.29	88.46 ± 1.17	83.14 ± 1.11	94.44 ± 1.82
K	53.97 ± 0.58	51.04 ± 1.03	89.62 ± 2.54	89.12 ± 0.26	98.33 ± 1.25
1	22.89 ± 1.09	1.33 ± 7.33	90.00 ± 1.03	30.21 ± 30.0	93.77 ± 11.6
2	45.46 ± 5.00	41.53 ± 3.04	87.35 ± 3.80	81.79 ± 0.26	90.38 ± 4.87
3	26.69 ± 2.61	30.91 ± 8.28	87.18 ± 7.24	75.93 ± 1.26	90.06 ± 4.53
4	22.79 ± 9.7	21.17 ± 3.25	83.17 ± 5.52	89.11 ± 1.12	96.84 ± 4.89
5	37.71 ± 6.67	69.79 ± 2.13	86.75 ± 2.55	79.28 ± 1.34	95.65 ± 1.95
6	89.57 ± 1.71	91.78 ± 0.78	89.08 ± 3.06	92.82 ± 0.32	96.95 ± 0.96
7	39.54 ± 11.4	19.85 ± 7.59	69.89 ± 29.7	39.69 ± 2.13	91.48 ± 24.0
8	87.46 ± 2.15	87.84 ± 3.15	85.25 ± 2.40	99.22 ± 0.31	89.11 ± 3.09
9	47.78 ± 19.04	0.00 ± 0.00	49.0 ± 49.0	19.00 ± 2.05	92.72 ± 8.38
10	49.46 ± 1.91	52.53 ± 1.24	86.47 ± 7.69	74.28 ± 0.89	92.40 ± 2.86
11	70.89 ± 2.49	61.88 ± 4.33	91.88 ± 5.03	91.12 ± 0.25	93.97 ± 3.29
12	37.14 ± 5.56	37.46 ± 3.85	77.82 ± 5.16	85.88 ± 2.37	87.56 ± 3.44
13	32.68 ± 7.17	85.02 ± 1.22	96.26 ± 5.29	50.86 ± 3.56	98.89 ± 0.87
14	81.32 ± 8.95	89.94 ± 2.59	89.16 ± 2.22	93.89 ± 1.37	96.89 ± 2.57
15	45.75 ± 5.12	44.64 ± 4.56	94.00 ± 7.49	93.76 ± 1.98	89.74 ± 2.65
16	29.60 ± 34.12	95.38 ± 1.94	99.89 ± 3.86	98.11 ± 2.67	96.89 ± 4.98