Research Article
HMFT: Hyperspectral and Multispectral Image Fusion Super-Resolution Method Based on Efficient Transformer and Spatial-Spectral Attention Mechanism
Table 6
QIs of the results by different methods on specified 4th test images from the Chikusei dataset.
| 4th (512 × 512 × 31) | Method | PSNR | SAM | ERGAS | RMSE | CC | SSIM |
| FUSE | 39.88 | 3.28 | 5.53 | 3.6121 | 0.9305 | 0.9202 | GLP-HS | 42.36 | 2.76 | 4.03 | 2.9853 | 0.9567 | 0.9403 | CSTF | 46.10 | 2.39 | 2.98 | 2.3787 | 0.9711 | 0.9566 | MHFnet | 45.74 | 1.91 | 2.95 | 0.0085 | 0.9744 | 0.9662 | HSRnet | 46.29 | 1.65 | 2.67 | 0.0084 | 0.9782 | 0.9683 | Ours | 46.75 | 1.55 | 2.50 | 0.0081 | 0.9801 | 0.9701 | Best value | +∞ | 0 | 0 | 0 | 1 | 1 |
|
|
The best values are underlined.
|