Research Article
HMFT: Hyperspectral and Multispectral Image Fusion Super-Resolution Method Based on Efficient Transformer and Spatial-Spectral Attention Mechanism
Table 5
Average QIs of the results for eight testing images on the Chikusei dataset.
| Method | PSNR | SAM | ERGAS | RMSE | CC | SSIM |
| FUSE | 38.88 | 3.47 | 5.78 | 3.7769 | 0.9286 | 0.9157 | GLP-HS | 41.02 | 3.20 | 4.41 | 3.2109 | 0.9519 | 0.9313 | CSTF | 44.40 | 3.14 | 3.45 | 3.0527 | 0.9592 | 0.9324 | MHFnet | 44.48 | 4.74 | 7.34 | 0.0066 | 0.9583 | 0.9771 | HSRnet | 44.82 | 1.80 | 2.93 | 0.0090 | 0.9758 | 0.9653 | Ours | 45.30 | 1.71 | 2.78 | 0.0087 | 0.9779 | 0.9874 | Best value | +∞ | 0 | 0 | 0 | 1 | 1 |
|
|
The best values are underlined.
|