Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Classification on Data set 2 (Combined-Feature Vectors)
| Macro F1 (%) | Weighted F1 (%) | Accuracy (%) | |
|---|---|---|---|
| Decision Tree | 98.97 | 99.01 | 99.01 |
| Random Forest | 99.74 | 99.75 | 99.75 |
| Bagging | 99.49 | 99.50 | 99.50 |
| Gradient Boosting | 99.49 | 99.50 | 99.50 |
| SVM | 94.12 | 94.31 | 94.30 |
| AdaBoost | 92.22 | 92.46 | 92.44 |
Image-based Classification with DGNSS-derived vector (Data set 1)
| Macro F1 (%) | Weighted F1 (%) | Accuracy (%) | |
|---|---|---|---|
| HOG + SVM | 68.90 | 83.43 | 85.34 |
| CNN | 71.08 | 85.17 | 86.77 |
Classification on Data set 1 (Direct Vector Classification)
| Macro F1 (%) | Weighted F1 (%) | Accuracy (%) | |
|---|---|---|---|
| Decision Tree | 56.34 | 74.72 | 72.24 |
| Random Forest | 63.36 | 79.96 | 78.81 |
| Bagging | 63.01 | 79.54 | 78.19 |
| Gradient Boosting | 63.01 | 79.06 | 77.32 |
| SVM | 48.21 | 58.95 | 52.79 |
| Ada Boosting | 52.79 | 69.23 | 64.56 |