Research Article

Sparse Dictionary-Based Magnetic Resonance Superresolution Imaging with Joint Loss Function Learning

Table 2

The PSNR/SSIM values under different dictionary blocks (superresolution ratio 1 : 4).

Dictionary block sizeSample set 1Sample set 2Sample set 3Sample set 4Sample set 5

5 × 529.82/0.875525.34/0.791329.16/0.875827.09/0.817227.54/0.8201
6 × 630.00/0.877825.49/0.794829.38/0.878227.22/0.831527.72/0.8285
7 × 730.30/0.881325.61/0.799629.49/0.879927.18/0.824327.91/0.8366
8 × 830.40/0.885725.71/0.810729.72/0.882327.23/0.836628.02/0.8434
9 × 930.43/0.890225.69/0.823829.78/0.890427.27/0.843028.05/0.8541
10 × 1030.75/0.900825.75/0.832329.94/0.898727.32/0.850328.11/0.8635
11 × 1130.67/0.891225.77/0.828529.90/0.892227.22/0.835428.12/0.8634
12 × 1230.53/0.890525.71/0.827429.82/0.891327.19/0.825527.93/0.8413
13 × 1330.31/0.882125.60/0.800229.64/0.881027.08/0.814127.78/0.8224