Table 5 Comparison with the state-of-the-art methods on BRACOL dataset.
From: Ambiguity-aware semi-supervised learning for leaf disease classification
Method | Accuracy | Precision | Recall | \(\hbox {F}_1\) |
|---|---|---|---|---|
Co-occurrence Matrix (GLCM)34 | 56.50 | 56.50 | 56.50 | 56.50 |
Local Binary Patterns (LBP)34 | 84.75 | 84.75 | 84.75 | 84.75 |
Resnet5035 | 97.07 | 96.85 | 96.99 | – |
DeiT28 | 96.78 | 96.58 | 96.66 | 96.60 |
Early Ensemble16 | 97.80 | 97.45 | 97.92 | – |
Late Ensemble16 | 97.80 | 97.54 | 97.70 | – |
The proposed method (DeiT with 50% label data) | 96.58 | 96.17 | 96.33 | 96.23 |
