Table 2 Comparison of different models on the self-built dataset
From: DWTFormer: a frequency-spatial features fusion model for tomato leaf disease identification
Models | Accuracy/% | Precision/% | Recall/% | F1-score/% | Params/M | FLOPs/G |
|---|---|---|---|---|---|---|
MobileNet V3-Large | 89.58 | 90.51 | 89.55 | 89.76 | 5.40 | 6.19 |
PVTv2-B/3 | 85.13 | 85.29 | 85.14 | 85.13 | 45.20 | 6.90 |
Swin Transformer-Base | 84.01 | 84.32 | 84.05 | 84.16 | 88.00 | 15.40 |
Vit-Base | 80.85 | 81.04 | 80.91 | 80.88 | 86.00 | 55.40 |
Mixer-B/16 | 80.15 | 80.82 | 80.22 | 80.17 | 59.00 | 11.60 |
DWTFormer | 97.63 | 98.01 | 97.22 | 96.84 | 33.69 | 7.96 |