Research Article
Sparse Dictionary-Based Magnetic Resonance Superresolution Imaging with Joint Loss Function Learning
Table 1
The PSNR/SSIM values using different block constructed dictionaries (superresolution 1 : 2).
| Dictionary block size | Sample set 1 | Sample set 2 | Sample set 3 | Sample set 4 | Sample set 5 |
| 5 × 5 | 32.24/0.9407 | 28.87/0.8996 | 32.04/0.9302 | 30.59/0.9145 | 30.34/0.9208 | 6 × 6 | 32.30/0.9412 | 28.87/0.8984 | 32.06/0.9309 | 30.53/0.9060 | 30.31/0.9084 | 7 × 7 | 31.99/0.9328 | 28.75/0.8875 | 31.95/0.9251 | 30.32/0.9025 | 30.14/0.9003 | 8 × 8 | 31.62/0.9297 | 28.44/0.8716 | 31.65/0.9195 | 30.09/0.8894 | 29.89/0.8936 | 9 × 9 | 31.20/0.9158 | 28.21/0.8658 | 31.33/0.9102 | 29.80/0.8721 | 29.52/0.8835 | 10 × 10 | 30.92/0.9021 | 27.95/0.8513 | 31.18/0.9031 | 29.60/0.8684 | 29.27/0.8748 |
|
|