Denoising on COVID-19 t=200
0.1367MSE (sigma=0.5)I-RP-GFRFT
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| I-RP-GFRFTRotation Matrix Type=roll, k=5/5/52025.11 | 0.1367 | 0.3308 | 0.4797 | |
| FFT soft threshold2025.11 | 0.2706 | 0.2756 | 0.2756 | |
| 1D total variation2025.11 | 0.4388 | 0.4718 | 1.2929 | |
| Tikhonov 1st diff.2025.11 | 0.9591 | 1.3972 | 6.6473 | |
| Moving average2025.11 | 1.0156 | 1.3714 | 6.1877 | |
| Savitzky–Golay2025.11 | 1.2357 | 2.3296 | 15.491 | |
| Median filter2025.11 | 1.4155 | 4.6757 | 11.191 | |
| Gaussian smoothing2025.11 | 2.1238 | 4.8701 | 17.989 | |
| ChebNet2025.11 | 111.21 | 470.84 | 1,353.1 | |
| GCNII2025.11 | 111.87 | 487.75 | 1,346.8 | |
| GCN2025.11 | 112.49 | 487.68 | 1,365.5 | |
| GAT2025.11 | 113.2 | 472.29 | 1,267 | |
| GraphSAGE2025.11 | 113.41 | 476.17 | 1,303.4 | |
| APPNP2025.11 | 121.05 | 464.39 | 1,350 |