Table 2 Accuracy assessment of three data-fusion methods applied to a complex and heterogeneous study area (Fig. 5). The units are reflectance (band1: blue band; band2: green band; band3: red band; band4: NIR band; band5: SWIR1 band; band6: SWIR2 band; R: correlation coefficient; RMSE: root mean square error; AD: average difference from the true reflectance; SSIM: structural similarity).
From: Improving spatiotemporal data fusion method in multiband images by distributing variates
R | RMSE | AD | SSIM | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
UBDF | FSDAF | RDSFM | UBDF | FSDAF | RDSFM | UBDF | FSDAF | RDSFM | UBDF | FSDAF | RDSFM | |
Band1 | 0.6770 | 0.7298 | 0.7353 | 0.0185 | 0.0177 | 0.0173 | 0.0041 | 0.0041 | 0.0039 | 0.9006 | 0.9094 | 0.9126 |
Band2 | 0.6461 | 0.7183 | 0.7319 | 0.0241 | 0.0232 | 0.0223 | 0.0023 | 0.0054 | 0.0051 | 0.8881 | 0.8965 | 0.8988 |
Band3 | 0.6630 | 0.7450 | 0.7600 | 0.0286 | 0.0269 | 0.0256 | 0.0015 | 0.0048 | 0.0043 | 0.8680 | 0.8814 | 0.8829 |
Band4 | 0.6920 | 0.8547 | 0.8637 | 0.0483 | 0.0359 | 0.0357 | 0.0096 | 0.0139 | 0.0130 | 0.7826 | 0.8382 | 0.8175 |
Band5 | 0.4423 | 0.8046 | 0.8096 | 0.0846 | 0.0544 | 0.0543 | 0.0340 | 0.0211 | 0.0196 | 0.6210 | 0.7142 | 0.7877 |
Band6 | 0.5084 | 0.8034 | 0.8074 | 0.0634 | 0.0456 | 0.0442 | 0.0210 | 0.0166 | 0.0156 | 0.6272 | 0.7272 | 0.7150 |
