Research Article
[Retracted] Comparative Analysis of Deepfake Image Detection Method Using Convolutional Neural Network
Table 1
Obtained results after implementing the models.
| CNN architecture’s name | Accuracy | Precision | Recall | F1-score | AUC |
| VGG19 | 0.94 | 0.91 | 0.97 | 0.94 | 0.987 | VGG16 | 0.92 | 0.93 | 0.92 | 0.92 | 0.977 | VGGFace | 0.99 | 0.99 | 0.98 | 0.99 | 0.998 | DenseNet169 | 0.95 | 0.99 | 0.92 | 0.95 | 0.996 | DenseNet201 | 0.96 | 0.96 | 0.97 | 0.96 | 0.994 | DenseNet121 | 0.97 | 0.99 | 0.70 | 0.82 | 0.971 | ResNet50 | 0.97 | 0.99 | 0.95 | 0.97 | 0.997 | Custom model | 0.90 | 0.84 | 0.99 | 0.91 | 0.989 |
|
|