Table 2 Training parameters of models/detector.
Model / Detector | Training dataset | Transfer learning | Leaf detector | Learning rate | Momentum | Weight decay | Batch size | Input size | Step size | Epochs / Iters |
|---|---|---|---|---|---|---|---|---|---|---|
[1]VGG16 [1]AlexNet [1]GoogleNet [1]ResNet18 [1]ResNet50 [1]Wide_ResNet50_v2 [1]ResNet101 | Laboratory Laboratory Laboratory Laboratory Laboratory Laboratory Laboratory | / / / / / / / | / / / / / / / | 0.001 0.001 0.001 0.001 0.001 0.001 0.001 | 0.9 0.9 0.9 0.9 0.9 0.9 0.9 | / / / / / / / | 16 16 16 16 16 16 16 | 224 × 224 224 × 224 299 × 299 224 × 224 224 × 224 224 × 224 224 × 224 | 4 4 4 4 4 4 4 | 50 50 50 50 50 50 50 |
[2]ResNet50 [2]ResNet50 | Laboratory Laboratory | one-stage two-stage | / / | 0.001 0.001 | 0.9 0.9 | / / | 16 16 | 224 × 224 224 × 224 | 4 4 | 100 100 |
[3]ResNet50 | Laboratory | / | / | 0.001 | 0.9 | / | 16 | 224 × 224 | 4 | 50 |
[4]LS-RCNN [4]ResNet50 [4]ResNet50 | Nature Nature Nature | / / / | / / LS-RCNN | 0.001 0.001 0.001 | 0.9 0.9 0.9 | 0.0005 / / | 256 16 16 | 375 × 500* 224 × 224 224 × 224 | 9000 4 4 | 15,000 100 100 |
[5]VGG16 [5]AlexNet [5]GoogleNet [5]GoogleNet* [5]OurModel | Nature Nature Nature Nature Nature | / / / one-stage two-stage | / / / / LS-RCNN | 0.001 0.001 0.001 0.001 0.001 | 0.9 0.9 0.9 / 0.9 | / / / 1e-3 / | 50 50 50 50 50 | 224 × 224 224 × 224 299 × 299 299 × 299 224 × 224 | 4 4 4 7 4 | 100 100 100 100 100 |
