Optics and Precision Engineering, Volume. 30, Issue 3, 320(2022)
Infrared and visible image fusion based on WEMD and generative adversarial network reconstruction
Figures & Tables(8)
Fig. 2. Flow chart of algorithm based on window empirical mode decomposition
Fig. 5. Comparison of experimental results of different image fusion methods
Table 1. Objective evaluation index of image fusion contrast experiment
View tableView in ArticleTable 1. Objective evaluation index of image fusion contrast experiment
Image Fusion method AG EI EN SSIM MI Image1 ASR 6.089 8 49.085 9 6.182 4 0.77 0.378 6 EP 2.447 7 19.428 0 5.869 6 0.778 0 0.322 2 CSMCA 3.968 7 18.750 4 6.322 4 0.794 2 1.204 3 ECNN 4.576 8 20.485 7 6.459 8 0.845 7 1.125 4 GAN 3.806 0 13.930 3 5.758 1 0.800 0 0.138 5 Proposed 7.367 0 50.291 5 7.367 0 0.803 2 0.882 9 Image2 ASR 6.633 3 40.386 1 6.436 4 0.698 3 0.324 1 EP 2.870 4 23.452 0 5.607 6 0.695 9 0.882 5 CSMCA 4.768 0 19.434 4 6.519 1 0.710 6 0.709 9 ECNN 4.742 1 18.867 5 6.482 1 0.712 4 0.795 6 GAN 3.016 9 13.523 7 4.562 8 0.758 9 0.114 Proposed 7.388 4 55.040 2 7.385 7 0.661 8 0.896 9 Image3 ASR 3.547 6 22.188 7 5.947 8 0.938 6 0.262 4 EP 5.295 7 34.146 0 6.766 8 0.782 0 0.616 2 CSMCA 2.352 1 15.846 2 6.388 3 0.848 2 0.440 7 ECNN 2.185 4 19.357 8 6.258 4 0.854 6 0.798 4 GAN 4.321 3 18.896 6.713 1 0.824 1 0.280 8 Proposed 6.325 4 40.236 4 7.112 3 0.845 6 0.812 7 Image4 ASR 3.787 4 33.006 3 6.296 3 0.811 1 0.294 6 EP 6.632 0 34.297 0 6.979 9 0.558 4 0.453 5 CSMCA 4.031 8 17.754 1 6.686 2 0.687 9 0.596 3 ECNN 2.324 8 21.386 5 6.945 8 0.724 8 0.865 4 GAN 4.197 1 14.892 1 7.069 4 0.812 2 0.227 5 Proposed 6.712 3 45.796 0 7.510 9 0.782 8 1.125 6 Image5 ASR 7.548 7 26.845 7 7.154 8 0.258 7 0.295 8 EP 7.857 4 38.311 7 6.808 5 0.727 2 0.363 4 CSMCA 4.707 5 18.262 7 6.690 2 0.673 1 0.275 1 ECNN 3.548 7 26.354 7 6.754 8 0.715 4 0.548 7 GAN 4.551 1 15.941 3 6.552 3 0.710 4 0.305 9 Proposed 7.195 5 54.262 0 7.223 0 0.737 5 0.791 5 Image6 ASR 4.759 9 30.117 8 6.607 1 0.770 2 0.280 5 EP 5.419 5 33.449 7 7.093 6 0.723 9 0.888 1 CSMCA 3.094 9 19.237 4 6.737 3 0.777 5 0.772 7 ECNN 3.214 8 18.657 8 6.985 4 0.754 6 0.854 7 GAN 3.439 7 17.430 9 7.076 9 0.785 5 0.297 7 Proposed 5.936 5 46.171 8 7.610 7 0.832 7 1.533 0
Table 2. Running time of each method on TNO dataset
View tableView in ArticleTable 2. Running time of each method on TNO dataset
Fusion algorithm Running time ASR 333.99 EP 0.13 CSMCA 405.69 ECNN 121.34 GAN 60.13 Ours 40.57
Tools
Get Citation
Copy Citation Text
Yanchun YANG, Xiaoyu GAO, Jianwu DANG, Yangping WANG. Infrared and visible image fusion based on WEMD and generative adversarial network reconstruction[J]. Optics and Precision Engineering, 2022, 30(3): 320
Paper Information
Category: Information Sciences
Received: May. 27, 2021
Accepted: --
Published Online: Mar. 4, 2022
The Author Email: Yanchun YANG (yangyanchun102@sina.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20