Optics and Precision Engineering, Volume. 32, Issue 10, 1567(2024)
Infrared image and visible image fusion algorithm based on secondary image decomposition
Figures & Tables(16)
Table 1. Comparison of fusion results of three fusion strategies in TNO dataset
View tableView in ArticleTable 1. Comparison of fusion results of three fusion strategies in TNO dataset
EN SD SF VIF AG SCD 像素相加法 7.338 9 9.843 0 0.046 8 0.828 0 4.529 4 1.927 8 加权相加法 6.742 4 9.116 2 0.027 7 0.698 0 2.725 8 1.695 2 通道注意力 7.337 1 9.816 4 0.047 0 0.812 6 4.445 7 1.864 4
Table 2. Comparison of fusion results of three fusion strategies in RoadScene dataset
View tableView in ArticleTable 2. Comparison of fusion results of three fusion strategies in RoadScene dataset
EN SD SF VIF AG SCD 像素相加法 7.377 8 10.929 6 0.055 6 0.715 4 5.452 3 1.809 2 加权相加法 6.920 3 9.715 2 0.035 6 0.651 9 3.615 5 1.459 2 通道注意力 7.357 3 11.247 6 0.056 0 0.690 7 5.303 4 1.796 3
Table 3. Ten fusion methods were objectively compared on the TNO dataset
View tableView in ArticleTable 3. Ten fusion methods were objectively compared on the TNO dataset
EN SD SF VIF AG SCD MS_SSIM Qabf DenseFuse 6.538 5 9.028 2 0.024 6 0.667 9 2.446 7 1.633 6 0.872 4 0.331 3 RFN-Nest 7.081 4 9.637 6 0.021 7 0.792 4 2.495 8 1.817 6 0.906 1 0.327 9 IFCNN 6.889 7 9.342 0 0.046 6 0.784 7 4.497 8 1.699 3 0.909 3 0.504 2 NestFuse 7.037 2 9.521 7 0.036 3 0.863 3 3.482 2 1.729 8 0.880 1 0.531 7 GTF 6.975 2 9.652 4 0.032 9 0.606 8 3.116 4 0.980 9 0.809 8 0.319 4 Dualbranch 6.523 4 8.989 3 0.023 0 0.661 9 2.377 4 1.624 4 0.867 6 0.321 2 LRRNet 7.160 0 9.317 3 0.035 6 0.785 9 3.617 0 1.618 8 0.866 4 0.358 6 SMOA 6.724 6 9.193 3 0.025 1 0.720 8 2.461 7 1.715 0 0.885 7 0.306 1 SDNet 6.722 8 9.505 8 0.035 6 0.966 2 3.393 6 0.880 7 0.726 7 0.428 9 SIDFuse 7.338 9 9.843 0 0.046 8 0.828 0 4.529 4 1.907 8 0.931 2 0.432 9
Table 4. Ten fusion methods were objectively compared on RoadScene dataset
View tableView in ArticleTable 4. Ten fusion methods were objectively compared on RoadScene dataset
EN SD SF VIF AG SCD MS_SSIM Qabf GTF 7.394 8 10.474 5 0.037 3 0.573 6 3.120 9 1.009 7 0.732 3 0.337 6 DenseFuse 6.802 5 9.403 3 0.036 9 0.638 8 3.708 2 1.391 5 0.848 1 0.353 1 RFN-Nest 7.299 2 9.958 8 0.030 7 0.691 2 3.389 4 1.727 2 0.863 9 0.292 6 IFCNN 6.911 6 9.470 3 0.054 9 0.639 3 5.228 1 1.439 0 0.900 1 0.521 8 NestFuse 7.396 5 10.329 6 0.053 1 0.824 4 5.075 3 1.707 1 0.865 0 0.502 2 Dualbranch 6.784 6 9.388 7 0.032 7 0.640 1 3.362 6 1.386 2 0.856 5 0.418 0 LRRnet 7.132 8 9.926 8 0.049 0 0.673 7 4.718 3 1.594 7 0.788 8 0.348 1 SMOA 6.957 6 9.641 1 0.041 2 0.672 9 4.093 5 1.539 2 0.877 0 0.432 2 SDNet 7.232 1 9.813 3 0.058 3 0.737 5 5.994 5 1.444 8 0.893 1 0.495 3 SIDFuse 7.487 6 10.796 4 0.060 1 0.728 2 5.969 9 1.837 1 0.911 7 0.488 4
Table 5. TNO data set ablation experiment
View tableView in ArticleTable 5. TNO data set ablation experiment
EN SD SF VIF AG SCD 实验a 7.004 3 8.876 3 0.040 7 0.766 4 4.407 4 1.797 6 实验b 7.032 9 8.964 1 0.044 7 0.776 2 4.175 1 1.815 0 实验c 7.220 4 9.472 5 0.045 4 0.816 0 4.362 0 1.875 8 实验d 7.256 2 9.715 2 0.047 4 0.821 6 4.801 9 1.901 4 实验e 7.338 9 9.843 0 0.046 8 0.828 0 4.529 4 1.907 8
Tools
Get Citation
Copy Citation Text
Xin MA, Chunyu YU, Yixin TONG, Jun ZHANG. Infrared image and visible image fusion algorithm based on secondary image decomposition[J]. Optics and Precision Engineering, 2024, 32(10): 1567
Paper Information
Category:
Received: Sep. 26, 2023
Accepted: --
Published Online: Jul. 8, 2024
The Author Email: YU Chunyu (yucy@njupt.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20