Table 3.
Performance metrics of the models on Sugarcane dataset.
| Model | Accuracy | Precision | Recall | F1-Score | Model | Accuracy | Precision | Recall | F1-Score |
|---|---|---|---|---|---|---|---|---|---|
| ResNet5025 | 84.00 | 84.67 | 83.76 | 83.74 | GoogLeNet21 | 82.00 | 81.94 | 82.06 | 81.95 |
| DenseNet20125 | 89.80 | 89.91 | 89.73 | 89.82 | ViT-L3229 | 87.60 | 87.54 | 87.59 | 87.50 |
| ConvNeXt-base26 | 90.60 | 89.67 | 90.35 | 89.78 | ViT-B3229 | 88.80 | 88.90 | 88.71 | 88.67 |
| ShuffleNetV2 x2.027 | 89.60 | 90.80 | 89.47 | 90.13 | Swin V2B12 | 88.40 | 88.48 | 88.32 | 88.35 |
| SqueezeNet 1.027 | 86.00 | 86.28 | 85.91 | 85.88 | Swin V2S12 | 87.60 | 87.92 | 87.61 | 87.66 |
| MobileNetV225 | 87.70 | 88.92 | 87.88 | 88.12 | Mob-Res (no finetune) | 88.00 | 89.96 | 87.87 | 87.90 |
| MobileNetV3Large25 | 87.68 | 88.85 | 86.51 | 87.76 | Mob-Res (finetune) | 85.60 | 85.76 | 85.50 | 85.53 |
| ResNeXt50 32x4d26 | 86.40 | 86.33 | 86.34 | 86.30 |