Table 2 Results obtained from the proposed method and comparative methods.
From: Using ensemble learning for classifying artistic styles in traditional Chinese woodcuts
Methods | Precision | Recall | F-measure | Accuracy |
|---|---|---|---|---|
Proposed | 0.9372 | 0.9367 | 0.9367 | 93.67% |
CNN1 + CNN2 | 0.8562 | 0.8544 | 0.8547 | 85.44% |
CNN2 + CNN3 | 0.8399 | 0.8378 | 0.8381 | 83.78% |
Mohammadi et al. 12 | 0.8633 | 0.8622 | 0.8624 | 86.22% |
Yang 13 | 0.8222 | 0.8211 | 0.8209 | 82.11% |
Zhao et al. 14 | 0.8904 | 0.8889 | 0.889 | 88.89% |
Fine-tuned ResNet-50 | 0.9041 | 0.9022 | 0.9023 | 90.22% |
Fine-tuned ViT-B/16 | 0.9195 | 0.9189 | 0.9188 | 91.89% |
