Table 3.
Classification accuracies for the original spatial resolution (0.07 m) UAV image classified using the various classification algorithms available in GEE.
| Accuracy Assessment | ||||||||
|---|---|---|---|---|---|---|---|---|
| Classification algorithm | CART | SVM | RF | GTB | ||||
| Overall Accuracy (%) | 81.00 | 83.00 | 86.00 | 85.00 | ||||
| Kappa Coefficient |
0.77 |
0.79 |
0.82 |
0.81 |
||||
| PA (%) |
UA (%) |
PA (%) |
UA (%) |
PA (%) |
UA (%) |
PA (%) |
UA (%) |
|
| Buildings and infrastructure | 98.00 | 88.00 | 88.00 | 100.00 | 98.00 | 93.00 | 98.00 | 91.00 |
| Bareground | 85.00 | 100.00 | 100.00 | 89.00 | 93.00 | 97.00 | 88.00 | 97.00 |
| Crops | 80.00 | 69.00 | 67.00 | 77.00 | 78.00 | 78.00 | 84.00 | 79.00 |
| Grassland | 69.00 | 79.00 | 78.00 | 73.00 | 80.00 | 80.00 | 76.00 | 80.00 |
| Trees & shrubs | 76.00 | 76.00 | 86.00 | 80.00 | 81.00 | 81.00 | 81.00 | 79.00 |