Research Article
Using Deep Convolutional Neural Networks for Image-Based Diagnosis of Nutrient Deficiencies in Rice
Table 6
Recall, precision, and f1-score per class of nutrient deficiency for DenseNet121.
| Class | Recall | Precision | f1 | Samples |
| Full | 0.96 | 0.90 | 0.93 | 58 | −N | 0.98 | 1.00 | 0.99 | 92 | −P | 1.00 | 0.97 | 0.98 | 61 | −K | 0.98 | 1.00 | 0.99 | 43 | −Ca | 0.95 | 0.95 | 0.95 | 42 | −Mg | 1.00 | 1.00 | 1.00 | 43 | −S | 1.00 | 1.00 | 1.00 | 78 | −Mn | 0.98 | 0.96 | 0.97 | 104 | −Fe | 0.83 | 0.94 | 0.88 | 31 | −Zn | 0.97 | 0.96 | 0.97 | 112 | −Si | 0.97 | 1.00 | 0.98 | 64 |
|
|