Table 2 Average performance of each model across folds for each threshold metric on the test set
| Threshold metrics (%) | |||||
|---|---|---|---|---|---|
| Model | Accuracy | Precision | Recall | F1-Score | |
| Feature extraction | VGG16 | 88.2 ± 1.2 | 88.4 ± 1.5 | 88.1 ± 0.9 | 87.8 ± 1.0 |
| InceptionV3 | 82.2 ± 1.8 | 83.7 ± 2.4 | 81.1 ± 2.2 | 81.7 ± 1.4 | |
| ResNet50V2 | 88.2 ± 1.1 | 88.6 ± 1.3 | 88.1 ± 07 | 87.9 ± 0.9 | |
| Xception | 90.1 ± 1.6 | 89.8 ± 0.9 | 90.7 ± 0.4 | 89.8 ± 0.7 | |