Research Article
Machine-Learning-Based Human Fall Detection Using Contact- and Noncontact-Based Sensors
Table 8
Summary of results of architectures implemented in case of optical flow images from the frontal view.
| Architecture used | Accuracy (%) | Precision (%) | Sensitivity (%) | Specificity (%) | F1-Score (%) |
| VGG-16 | 92.625 | 95.46 | 89.50 | 95.75 | 92.38 | Xception | 93.625 | 96.28 | 90.75 | 96.50 | 93.43 | DenseNet | 96.00 | 96.46 | 95.50 | 96.50 | 95.97 |
|
|