Research Article
A New Semisupervised-Entropy Framework of Hyperspectral Image Classification Based on Random Forest
Table 4
The comparison of overall accuracy and Kappa coefficient under different classifier algorithm of Hyperion data.
| Comparative classification method | Run time (seconds) | Overall Accuracy (%) | Kappa Coefficient | T1 | T2 | T3 | T1 | T2 | T3 | T1 | T2 | T3 |
| Minimum distance classification algorithm | 12 | 15 | 14 | 89.26 | 89.83 | 90.95 | 0.7567 | 0.7693 | 0.7952 | Support-vector machine classification algorithm | 25 | 27 | 26 | 94.55 | 94.63 | 94.91 | 0.8744 | 0.8752 | 0.8829 | Classification and regression trees (CART) | 7.5 | 7.2 | 6.7 | 86.0806 | 93.3795 | 94.1106 | 0.6806 | 0.8468 | 0.8901 | This-algorithm classifies (5% of the labeled labels) | 6.04 | 6.35 | 6.97 | 97.22 | 98.26 | 98.83 | 0.9495 | 0.9684 | 0.9788 | This-algorithm classifies(10% of labeled labels) | 7.16 | 7.69 | 8.16 | 98.20 | 98.90 | 99.17 | 0.9711 | 0.9820 | 0.9904 |
|
|