Table 3 Quantitative comparison with seven algorithms on the SPECT-MRI fusion task (averaged over 24 image pairs). Values in bold indicate the best performance and values in italic indicate the second-best.
From: A novel multimodel medical image fusion framework with edge enhancement and cross-scale transformer
Metrics | U2Fusion | EMFusion | SwinFusion | DeFusion | MuFusion | TIMFusion | CDDFuse | Proposed |
|---|---|---|---|---|---|---|---|---|
MI | 2.6466 | 3.0753 | 3.2801 | 2.9186 | 3.0064 | 3.8820 | 3.9318 | 4.5115 |
Qabf | 0.4781 | 0.6680 | 0.7131 | 0.4098 | 0.5896 | 0.6269 | 0.7303 | 0.7378 |
VIF | 0.4053 | 0.6300 | 0.7153 | 0.5003 | 0.5333 | 0.6133 | 0.8397 | 0.8796 |
SSIM | 0.2502 | 0.6223 | 0.2868 | 0.5823 | 0.2903 | 0.6116 | 0.6254 | 0.6152 |
Qcb | 0.3183 | 0.6492 | 0.3608 | 0.5780 | 0.3782 | 0.6211 | 0.6782 | 0.6779 |
Qcv\(\downarrow\) | 409.00 | 83.80 | 59.66 | 305.49 | 164.31 | 213.15 | 53.81 | 50.37 |
