Table 5 The average SNR, PSNR, SSIM and reconstruction time of each slice with a different kernel size in the main path in MFU and the final reconstruction layer at the \(10^5\) iteration
From: Fusing multi-scale information in convolution network for MR image super-resolution reconstruction
Network | SNR | PSNR | SSIM | Time (s) |
|---|---|---|---|---|
\(MFCN_{s_{713}}\) | 26.574 | 39.6064 | 0.9875 | 0.009123 |
\(MFCN_{BL}\) | 27.0741 | 40.1064 | 0.9891 | 0.019142 |
\(MFCN_{s_{957}}\) | 27.0415 | 40.0738 | 0.989 | 0.020472 |
\(MFCN_{s_{1159}}\) | 26.9858 | 40.0181 | 0.9885 | 0.027577 |