TABLE IV.
Ablation Study on Our MDPET in Terms of Denoising. ± LSTM Means MDPET With or Without BiConvLSTM and ± GAN Means MDPET With or Without Adversarial Learning.
| Method-mean(std) | NMAE | SSIM | PSNR |
|---|---|---|---|
| Ours-LSTM-GAN | .1058(.0140) | .9587(.0063) | 30.95(1.89) |
| Ours+LSTM-GAN | .0921(.0137) | .9613(.0062) | 31.64(1.88) |
| Ours+LSTM+GAN | .0883(.0133)* | .9669(.0054)* | 32.28(1.89)* |
Means the Difference to the Baseline (1st Row) Are Significant at p < 0.05