Table 13.
Comparison between macro average precision, sensitivity, f-measure and specificity for different models applied on MosMed dataset
| Models | Macro average | |||
|---|---|---|---|---|
| Prec (%) | Sens (%) | F-measure (%) | Spec (%) | |
| Spatial articulated feature (HOG + DAISY) fusion + SVM | 73.45 | 84.41 | 72.92 | 95.82 |
| Inception-v3 | 87.18 | 85.80 | 86.45 | 95.89 |
| Mobilenet-v2 | 93.17 | 88.88 | 90.90 | 96.48 |
| ResNet-50 | 92.80 | 89.02 | 90.80 | 96.54 |
| Xception | 94.42 | 91.30 | 92.78 | 97.30 |
| The proposed modified Norm-VGG16 | 97.46 | 94.22 | 95.77 | 98.42 |
| The proposed COV-CAF model | 97.58 | 96.73 | 97.14 | 98.97 |