Table 7 Comparative analysis of classification results on Chinese medicinal blossom, Chinese medicinal leaf, and fruit datasets.
Methods | Medicinal blossom | Medicinal leaf | Fruit dataset | |||
|---|---|---|---|---|---|---|
Top-1 ACC (%) | AUC (%) | Top-1 ACC (%) | AUC (%) | Top-1 ACC (%) | AUC (%) | |
VGG | 84.47 | 94.0 | 83.96 | 94.0 | 67.189 | 80.0 |
ResNet | 89.87 | 95.5 | 88.27 | 97.0 | 91.864 | 95.0 |
MobileNets | 96.78 | 99.0 | 98.4 | 99.9 | 78.25 | 85.0 |
DenseNet | 93.85 | 96.5 | 97.54 | 99.9 | 94.80 | 98.0 |
EffcientNet | 97.02 | 99.0 | 99.2 | 100 | 92.74 | 96.0 |
ViT | 90.75 | 95.5 | 93.72 | 96.5 | 78.93 | 85.0 |
CoAtNet | 93.38 | 96.0 | 97.38 | 99.9 | 90.86 | 95.0 |
FocalNet | 91.75 | 95.0 | 94.35 | 98.0 | 88.95 | 95.0 |
Swin Transformer | 95.58 | 98.0 | 97.58 | 99.9 | 94.85 | 98.0 |
CMT | 93.86 | 96.0 | 96.38 | 99.0 | 93.74 | 96.0 |
CvT | 93.54 | 96.0 | 96.36 | 99.0 | 92.85 | 96.0 |
PVT | 90.95 | 95.5 | 94.77 | 98.0 | 91.95 | 95.5 |
MaxViT | 92.35 | 96.0 | 96.7 | 99.0 | 92.63 | 96.0 |
EfficientViT | 96.46 | 99.0 | 99.31 | 100 | 95.05 | 99.0 |
SwinFG | 96.73 | 99.0 | 97.54 | 99.9 | 96.52 | 99.0 |
Ours | 97.34 | 100 | 99.73 | 100 | 97.87 | 100 |
