HMFT: Hyperspectral and Multispectral Image Fusion Super-Resolution Method Based on Efficient Transformer and Spatial-Spectral Attention Mechanism

Table 3

Average QIs for the ten test images on the Harvard data set.

Method	PSNR	SAM	ERGAS	RMSE	CC	SSIM
FUSE	37.88	3.33	5.73	4.3460	0.9817	0.9382
GLP-HS	47.82	2.71	2.37	1.2121	0.9960	0.9897
CSTF	45.72	9.95	3.20	1.5777	0.9921	0.9731
MHFnet	44.48	4.74	7.34	0.0066	0.9583	0.9771
HSRnet	48.63	2.66	2.34	0.0042	0.9963	0.9894
Ours	49.10	2.45	2.23	0.0040	0.9966	0.9977
Best value	+∞	0	0	0	1	1

The best values are underlined.