Table 3 Average accuracy (AA), overall accuracy (OA), and kappa for some segmentation-based and filter-based methods on Indian Pines and Pavia University data sets
Methods | Ref. | Indian Pines | PaviaU | ||||
|---|---|---|---|---|---|---|---|
AA | OA | Kappa | AA | OA | Kappa | ||
SVM-MSF | [22] | 91.06 | 88.4 | 87 | 84.14 | 92.3 | 80 |
MSSC-MSF | [23] | 92.3 | 94.2 | 91 | 98.6 | 97.9 | 97 |
RD-MSF | [24] | 93.73 | 91.33 | – | – | – | – |
Pixon based | [6] | 92.18 | 89.94 | 88 | 94.72 | 95.17 | 94 |
SVMMRF-HSRM | [5] | – | 93.10 | 92.11 | – | – | – |
MLRsubMLL-HSRM | [5] | – | 80.21 | 82.60 | – | – | – |
EPF | [25] | 94.46 | 90.28 | 89.98 | 98.38 | 99.09 | 98.77 |
MSF-KNN | [29] | 90.1 | 88.67 | 87.32 | 94.39 | 95.1 | 94.12 |
AMG-M-HSEG | [30] | 94.83 | 91.4 | 89.48 | 97.18 | 97.45 | 97.23 |
MSEPF-MMSF (Proposed method) | – | 95.64 | 91.45 | 90.32 | 98.36 | 98.8 | 98.41 |