Research Article
Application of Image Super-Resolution Reconstruction in Gymnastics Training by Using Internet of Things Technology
Table 2
Comparison of the features of traditional methods and SRCNN.
| Dataset | Super-resolution scale | Bicubic PSNR (dB) | NE-LLE PSNR (dB) | A + PSNR (dB) | SRCNN PSNR (dB) |
| Set5 | ×2 | 33.66 | 35.77 | 36.54 | 36.66 | Set14 | ×2 | 30.23 | 31.76 | 32.28 | 32.45 | B200 | ×2 | 28.38 | 29.67 | 30.14 | 30.29 | Set5 | ×3 | 30.39 | 31.84 | 32.59 | 32.75 | Set14 | ×3 | 27.54 | 28.60 | 29.13 | 29.30 | B200 | ×3 | 25.94 | 26.67 | 27.05 | 27.18 | Set5 | ×4 | 28.42 | 29.61 | 30.28 | 30.49 | Set14 | ×4 | 26.00 | 26.81 | 27.32 | 27.50 | B200 | ×4 | 24.65 | 25.21 | 25.51 | 25.60 |
|
|