Research Article
Median Filter Based Compressed Sensing Model with Application to MR Image Reconstruction
Table 1
Numerical results for the synthetic phantom image and real MR images with different sampling ratios.
| Image | Samp.Ratio | PSNR (dB) | Time (s) | Rerror |
| Phantom | 3.90% | 28.19 | 6.34 | 0.165 | 512×512 | 5.76% | 34.55 | 4.65 | 0.078 | | 9.50% | 42.49 | 3.08 | 0.040 |
| Image | Samp.Ratio | PSNR (dB) | Time (s) | Rerror |
| Brain 1 | 7.60% | 29.89 | 1.74 | 0.120 | 512×512 | 15.00% | 34.04 | 1.55 | 0.070 | | 35.01% | 37.47 | 1.55 | 0.051 |
| Image | Samp.Ratio | PSNR (dB) | Time (s) | Rerror |
| Brain 2 | 7.60% | 31.05 | 1.59 | 0.130 | 512×512 | 18.50% | 37.52 | 1.55 | 0.060 | | 35.00% | 44.46 | 1.53 | 0.020 |
|
|