Optics and Precision Engineering, Volume. 31, Issue 20, 3050(2023)
CT and PET medical image fusion based on LL-GG-LG Net
Figures & Tables(11)
Fig. 6. Subjective comparison of fused images by different methods
Table 1. Mean value of objective evaluation metrics.
View tableView in ArticleTable 1. Mean value of objective evaluation metrics.
No. Methods AG EI QAB/F SF SD IE 1 LATLRR 9.669 55.903 0 0.470 4 26.66 36.52 6.537 1 LRD 10.009 67.610 9 0.378 6 27.85 39.11 6.007 8 DPCN 8.568 65.025 6 0.372 7 24.31 41.52 6.057 2 Res2Net 4.860 34.026 1 0.188 4 13.34 33.01 5.565 5 DIFNet 5.676 44.391 2 0.271 6 17.92 23.78 5.834 2 EMFusion 9.487 75.529 0 0.509 9 30.03 40.46 6.254 3 DDcGAN 10.609 57.058 7 0.310 0 29.16 32.35 5.876 9 OURS 12.901 78.936 0 0.520 0 34.42 46.63 6.555 2 2 LATLRR 9.494 54.709 2 0.469 8 25.63 36.33 6.445 1 LRD 10.152 70.681 0 0.415 4 29.12 42.90 6.361 0 DPCN 9.143 70.115 4 0.437 3 24.80 39.65 6.384 3 Res2Net 3.568 34.822 2 0.188 1 13.74 33.13 5.657 4 DIFNet 5.445 43.107 7 0.263 9 17.83 24.84 5.679 1 EMFusion 9.225 70.558 6 0.512 3 29.14 40.73 6.080 6 DDcGAN 10.383 56.541 6 0.302 5 27.57 32.41 6.146 7 OURS 12.577 75.507 8 0.532 8 32.91 44.72 6.586 3 3 LATLRR 9.312 56.217 1 0.458 9 24.45 33.31 6.209 9 LRD 10.568 75.193 2 0.418 6 29.38 43.38 6.067 0 DPCN 9.234 70.159 1 0.416 1 24.76 42.98 6.656 2 Res2Net 4.845 32.287 0 0.195 1 13.11 34.59 5.190 9 DIFNet 6.593 45.018 0 0.263 7 19.95 26.04 5.224 6 EMFusion 10.421 73.309 5 0.483 3 29.24 42.49 6.325 1 DDcGAN 10.542 59.717 5 0.322 7 27.55 35.69 6.520 1 OURS 12.771 78.064 2 0.514 2 32.77 48.94 6.766 2 4 LATLRR 7.344 52.859 9 0.463 6 24.14 40.98 6.777 6 LRD 7.854 56.040 9 0.418 6 26.49 44.90 6.276 7 DPCN 7.461 57.857 8 0.416 4 23.82 46.99 6.251 3 Res2Net 4.735 33.297 2 0.186 9 13.26 34.48 5.921 4 DIFNet 5.805 41.799 6 0.262 8 17.73 26.50 6.163 3 EMFusion 9.236 65.856 7 0.503 3 29.90 44.49 7.095 4 DDcGAN 8.390 47.059 3 0.323 7 27.55 35.69 6.521 0 OURS 10.257 74.071 5 0.518 1 33.77 48.65 6.966 2 5 LATLRR 8.243 59.719 8 0.472 5 24.02 35.28 6.483 7 LRD 8.715 63.709 9 0.393 2 27.34 39.86 5.926 2 DPCN 8.040 62.557 3 0.407 1 23.81 41.66 5.923 7 Res2Net 5.340 36.794 4 0.176 0 14.08 26.97 5.596 6 DIFNet 6.960 48.410 1 0.280 5 19.41 22.82 5.754 2 EMFusion 9.773 71.245 6 0.489 6 27.87 39.52 6.325 8 DDcGAN 9.091 51.866 5 0.299 9 26.08 30.74 5.740 8 OURS 11.048 90.424 1 0.525 4 31.67 44.93 6.543 2
Table 2. Mean value of objective evaluation metrics for ablation experiments
View tableView in ArticleTable 2. Mean value of objective evaluation metrics for ablation experiments
No. LL Module GG Module LG Module AG EI QAB/F SF SD IE Number of attention layers Residual connection No
（Average weighted）
Yes Single Double No Yes 1 √ × × × × × 8.802 4 63.593 0 0.465 2 26.055 5 33.141 3 5.130 1 2 × √ × × × × 10.149 3 77.014 6 0.487 6 29.444 7 39.209 3 5.584 6 3 × × √ × × × 6.490 7 47.797 3 0.387 9 22.891 9 36.838 8 4.394 8 4 × × × √ × × 7.870 1 52.358 2 0.403 6 24.558 9 36.227 8 5.133 2 5 × √ × √ √ × 8.947 9 62.319 3 0.491 8 26.224 1 36.537 3 5.836 4 6 × √ × √ × √ 11.906 6 85.917 3 0.516 0 32.293 6 44.694 9 6.232 5
Tools
Get Citation
Copy Citation Text
Tao ZHOU, Xiangxiang ZHANG, Huiling LU, Qi LI, Qianru CHENG. CT and PET medical image fusion based on LL-GG-LG Net[J]. Optics and Precision Engineering, 2023, 31(20): 3050
Paper Information
Category: Information Sciences
Received: Mar. 31, 2023
Accepted: --
Published Online: Nov. 28, 2023
The Author Email: Xiangxiang ZHANG (zxx19990503@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20