Research Article
Spectral-Spatial Hyperspectral Image Semisupervised Classification by Fusing Maximum Noise Fraction and Adaptive Random Multigraphs
Table 7
Classification results of different techniques for the Salinas dataset (16 classes).
| No. | Pixon-based | PCA-SPCA-2D-SSA | R-VCANet | RN-FSC | iCapsNet | RPNet | MBFSDA | SS-MNF-ARMG |
| 1 | 99.90 | 96.30 | 96.95 | 99.26 | 93.41 | 98.01 | 96.95 | 97.11 | 2 | 99.38 | 88.94 | 99.06 | 93.21 | 67.16 | 99.19 | 97.65 | 98.98 | 3 | 96.20 | 92.09 | 97.31 | 97.87 | 86.49 | 96.91 | 99.60 | 100.00 | 4 | 91.61 | 90.23 | 99.46 | 85.50 | 90.36 | 99.21 | 100.00 | 86.21 | 5 | 86.89 | 88.94 | 97.35 | 84.81 | 90.39 | 98.36 | 96.30 | 88.67 | 6 | 88.79 | 96.08 | 99.87 | 99.35 | 99.02 | 99.60 | 99.14 | 89.47 | 7 | 97.32 | 91.36 | 95.26 | 96.65 | 95.07 | 96.42 | 93.55 | 97.65 | 8 | 87.20 | 96.44 | 64.01 | 66.24 | 82.12 | 76.01 | 78.33 | 96.66 | 9 | 99.84 | 85.97 | 97.79 | 97.34 | 98.63 | 96.58 | 97.81 | 100.00 | 10 | 46.34 | 86.38 | 81.38 | 82.66 | 93.13 | 84.66 | 94.23 | 87.22 | 11 | 52.90 | 94.17 | 93.17 | 73.96 | 60.09 | 96.25 | 97.38 | 88.30 | 12 | 30.10 | 80.49 | 96.48 | 96.84 | 94.78 | 97.87 | 99.27 | 97.18 | 13 | 89.74 | 96.17 | 98.70 | 85.39 | 49.51 | 93.56 | 98.69 | 86.42 | 14 | 84.21 | 99.30 | 90.35 | 90.39 | 76.76 | 97.76 | 75.70 | 95.79 | 15 | 94.79 | 86.68 | 56.92 | 65.85 | 49.76 | 66.26 | 80.38 | 95.36 | 16 | 92.09 | 98.17 | 90.79 | 86.89 | 96.89 | 97.57 | 99.89 | 100.00 | AA | 83.58 | 91.43 | 90.93 | 87.64 | 77.26 | 93.39 | 94.05 | 94.06 | OA | 87.15 | 91.73 | 93.56 | 93.12 | 82.72 | 88.16 | 90.96 | 93.73 | Kappa | 85.63 | 90.90 | 85.72 | 85.44 | 74.72 | 86.83 | 89.95 | 93.03 |
|
|
Bold values represent the best results among these methods.
|