Table 2 The performance measurement results for the traditional CNN architecture and the proposed model with CapsNet
From: An optimized capsule neural networks for tomato leaf disease classification
Class number | Ordinary architecture without CapsNet | Proposed model with CapsNet | ||||||
|---|---|---|---|---|---|---|---|---|
Accuracy | Precision | Recall | F1-Score | Accuracy | Precision | Recall | F1-Score | |
1 | 92.00 | 82.60 | 92.30 | 87.18 | 97.00 | 82.40 | 96.60 | 88.94 |
2 | 62.00 | 68.00 | 61.60 | 64.64 | 59.00 | 73.10 | 58.80 | 65.17 |
3 | 78.00 | 80.90 | 77.60 | 79.22 | 74.00 | 91.70 | 74.40 | 82.15 |
4 | 74.00 | 82.70 | 74.50 | 78.39 | 69.00 | 96.70 | 69.30 | 80.74 |
5 | 69.00 | 82.00 | 69.00 | 74.94 | 88.00 | 81.70 | 88.10 | 84.78 |
6 | 81.00 | 78.00 | 81.10 | 79.52 | 90.00 | 84.70 | 90.40 | 87.46 |
7 | 84.00 | 75.00 | 83.90 | 79.20 | 83.00 | 82.90 | 82.90 | 82.90 |
8 | 96.00 | 95.80 | 95.50 | 95.65 | 98.00 | 95.60 | 97.80 | 96.69 |
9 | 79.00 | 95.90 | 79.10 | 86.69 | 86.00 | 95.50 | 85.80 | 90.39 |
10 | 98.00 | 89.90 | 98.10 | 93.82 | 99.00 | 91.80 | 98.80 | 95.17 |