Table 7.
Comparison of the proposed method with other recent literature.
| Authors., Ref., Year | Methods | Dataset size | Number of Classes | Category | Accuracy |
|---|---|---|---|---|---|
| Ruby et al. (2024) | Modified ResNet50 | 4500 | 4 | Wheat | 98.44% |
| Shoaib et al. (2022) | Modified U-Net | 18161 | 10 | Tomato | 99.35% |
| Xiang et al. (2025) | DWTFormer | 54306 | 9 | Tomato | 99.28% |
| Bhavani (2025) | Convolutional autoencoder | 1166 | 5 | Soybean | 92% |
| Seelwal et al. (2024) | Deep learning | 5932 | 6 | Rice | 94.25% |
| Alkanan and Gulzar (2024) | MobileNetV2 | 17,801 | 4 | Corn | 96% |
| Gulzar (2024) | Improved Inceptionv3 | 5513 | 5 | Soybean | 98.73% |
| Gulzar et al. (2025a) | Transfer learning | 1214 | 3 | Alfalfa | 99.45% |
| Gulzar and Ünal (2025b) | PL-DenseNet | 3505 | 4 | Pear | 99.18% |
| Gulzar and Ünal (2025c) | PlmNet | 400 | 3 | Plums | 97.58% |
| Proposed. (2025) | DSA-Net | 7915 | 5 | Pea | 99.12% |