Table 3 The classification evaluation outcomes for each pre-trained deep learning model using the Arabic Char 4k OCR dataset are compared to the proposed ensemble and hybrid transformer models.
The models | ACC | REC | PRE | F1 Score | AUC |
|---|---|---|---|---|---|
VGG16 | 98.41 | 98.40 | 98.81 | 98.41 | 99.20 |
ResNet50 | 97.94 | 97.90 | 98.60 | 97.89 | 98.90 |
Ensemble model | 98.72 | 98.72 | 98.98 | 98.74 | 99.30 |
The proposed hybrid transformer model | 99.51 | 99.55 | 99.49 | 99.53 | 99.20 |
