Table 2.
Summary of all the parameters and hyperparameters used for each model.
| Model | Loss function | Optimizer | Learning rate | Epochs | Batch size | ||
|---|---|---|---|---|---|---|---|
| Train | Test | Validate | |||||
| U-Net | Binary cross-entropy | Adam | 0.01 | 500 | 60 | 20 | 20 |
| ResUNet | Binary cross-entropy | Adam | 0.01 | 350 | 60 | 20 | 20 |
| Dense UNet | Binary cross-entropy | Adam | 0.01 | 250 | 60 | 20 | 20 |
| DUNet | Binary cross-entropy | Adam | 0.01 | 200 | 60 | 20 | 20 |
| Attention U-Net | Binary cross-entropy | Adam | 0.01 | 150 | 60 | 20 | 20 |
| UNet++ | Binary cross-entropy | Adam | 0.01 | 245 | 60 | 20 | 20 |
| MultiResUNet | Binary cross-entropy | Adam | 0.01 | 300 | 60 | 20 | 20 |
| RAUNet | Binary cross-entropy | Adam | 0.01 | 200 | 60 | 20 | 20 |
| Inception U-Net | Binary cross-entropy | Adam | 0.01 | 85 | 60 | 20 | 20 |
| U-Net GAN | Binary cross-entropy | Adam | 0.01 | 1150 | 60 | 20 | 20 |
| V-Net | Binary cross-entropy | Adam | 0.01 | 600 | 60 | 20 | 20 |
| W-Net | Binary cross-entropy | Adam | 0.01 | 120 | 60 | 20 | 20 |
| DeepLabV3+ | Binary cross-entropy | Adam | 0.01 | 10 | 60 | 20 | 20 |