| 模型 | 评估指标(平均值) | ||
| Accuracy | Precision | F1-score | |
| AlexNet | 85.39 | 84.19 | 83.97 |
| VGG16 | 84.90 | 82.35 | 83.37 |
| ResNet50 | 85.63 | 83.44 | 83.81 |
| ARL14 | 86.84 | 86.05 | 85.77 |
| ARL50 | 84.06 | 83.77 | 83.64 |
| 提出模型 | 88.29 | 87.37 | 87.47 |