Remote Sensing Image Fusion Algorithm Based on Two-Stream Fusion Network and Residual Channel Attention Mechanism

Table 2

Evaluation indicators of different remote sensing image fusion algorithms under WorldView-2 satellite datasets.

Reference comparison
Methods	SCC	SSIM	Q	PSNR	ERGAS	SAM
Brovey	0.8481	0.7095	0.5476	43.1899	5.8498	0.1004
PCA	0.8903	0.7031	0.4028	44.2140	4.7886	0.1069
IHS	0.8732	0.6831	0.4642	44.2607	5.0181	0.0951
SFIM	0.9139	0.7472	0.6146	44.6207	3.8994	0.0937
MTF_GTP	0.8858	0.6751	0.4254	44.8155	4.8367	0.0867
CNMF	0.9271	0.7706	0.6139	45.7792	3.8437	0.0888
PNN	0.9281	0.7791	0.6252	45.9311	3.7992	0.0885
DPRNN	0.9312	0.7825	0.6371	46.6754	3.5910	0.0851
PanNet	0.9350	0.7838	0.6215	46.2696	3.5771	0.0816
Our algorithm	0.9422	0.7891	0.6432	47.5762	3.4545	0.0656
Nonreference comparison
Methods					QNR
Brovey	0.1002		0.1063		0.8041
PCA	0.0911		0.0401		0.8724
IHS	0.0897		0.1143		0.8063
SFIM	0.0649		0.1257		0.8175
MTF_GLP	0.0624		0.0764		0.8643
CNMF	0.0595		0.1687		0.7818
PNN	0.0523		0.1022		0.8508
DPRNN	0.0448		0.0540		0.9036
PanNet	0.0323		0.0274		0.9412
Our algorithm	0.0204		0.0128		0.9671