Research Article
Generative Adversarial Network-Based Edge-Preserving Superresolution Reconstruction of Infrared Images
Table 2
Method comparison: quantitative results of the SRCNN, ESPCN, SRGAN, ESRGAN, and ESRGAN+ methods and the proposed method on the Itir_v1_0 dataset.
| | CC | PSNR (dB) | SSIM | VIF | UIQI | TIME (s) |
| SRCNN | 0.7937 | 20.2460 | 0.8966 | 0.7079 | 0.8746 | 2.5706 | ESPCN | 0.8464 | 22.1708 | 0.9185 | 0.8194 | 0.9877 | 1.0020 | SRGAN | 0.8152 | 22.6732 | 0.9154 | 0.7944 | 0.8610 | 0.6642 | ESRGAN | 0.9146 | 30.3205 | 0.9608 | 0.7752 | 0.9166 | 0.0937 | ESRGAN+ | 0.9547 | 34.1024 | 0.9844 | 0.8497 | 0.9541 | 0.2558 | Ours | 0.9962 | 35.2865 | 0.9943 | 0.9064 | 0.9936 | 0.1059 |
|
|