Table 5 Comparison of the classification accuracies (ACC (%)) for leaf 8 (wounded) and leaf 13 (systemic) SWPs using three different augmentation methods
From: Early detection of dark-affected plant mechanical responses using enhanced electrical signals
Leaves | Time-domain (ACC) | Deriv_1st (ACC) | Integral (ACC) | ||||||
|---|---|---|---|---|---|---|---|---|---|
AAE | VAE | GAN | AAE | VAE | GAN | AAE | VAE | GAN | |
Leaf 8 | 82.44 | 79.51 | 84.39 | 95.61 | 89.75 | 67.32 | 95.61 | 89.26 | 83.90 |
80.98 | 82.43 | 81.95 | 90.24 | 87.31 | 69.93 | 84.88 | 85.85 | 86.83 | |
80.49 | 81.95 | 81.95 | 99.02 | 90.24 | 80.49 | 91.22 | 88.78 | 84.88 | |
Leaf 13 | 87.80 | 64.88 | 76.10 | 93.66 | 89.27 | 76.59 | 97.07 | 91.71 | 84.39 |
88.78 | 77.56 | 76.59 | 86.83 | 88.78 | 71.22 | 91.21 | 91.21 | 84.39 | |
88.78 | 84.39 | 80.49 | 99.02 | 90.73 | 70.73 | 96.10 | 93.65 | 83.90 | |