Research Article
Google Earth Engine for Advanced Land Cover Analysis from Landsat-8 Data with Spectral and Topographic Insights
Table 4
Quantitative results after postprocessing step for land cover classification.
| Method | Classes | Precision | Recall | F1 score | OA | Kappa |
| SVM + SB | Water | 0.9836 | 0.9871 | 0.9853 | 0.8552 | 0.8162 | High vegetation | 0.9253 | 0.9284 | 0.9269 | Low vegetation | 0.6957 | 0.6971 | 0.6964 | Crops | 0.7217 | 0.7103 | 0.7159 | Built-up | 0.8581 | 0.8613 | 0.8597 |
| SVM + SB + SI | Water | 0.9928 | 0.9844 | 0.9886 | 0.9070 | 0.8817 | High vegetation | 0.9733 | 0.9116 | 0.9414 | Low vegetation | 0.7956 | 0.7866 | 0.7911 | Crops | 0.798 | 0.8609 | 0.8282 | Built-up | 0.9029 | 0.9416 | 0.9219 |
| SVM + SB + SI + TF | Water | 0.995 | 0.9925 | 0.9938 | 0.9290 | 0.9099 | High vegetation | 0.9753 | 0.9488 | 0.9619 | Low vegetation | 0.8881 | 0.8111 | 0.8479 | Crops | 0.8314 | 0.8864 | 0.858 | Built-up | 0.9159 | 0.9588 | 0.9368 |
|
|