Optics and Precision Engineering, Volume. 30, Issue 18, 2267(2022)
Two-stage image restoration using improved atmospheric scattering model
Figures & Tables(12)
Fig. 2. Dehazing results of different algorithms on the RESIDE dataset
Fig. 3. Dehazing results of different algorithms on the SOTS dataset
Fig. 4. Dehazing results of different algorithms on the O-HAZE dataset
Fig. 5. Dehazing results of different algorithms on the I-HAZE dataset
Fig. 6. Underwater image restoration results of different algorithms on the UIEB dataset
Table 1. Performance metrics of dehazing results of various dehazing algorithms corresponding to figure 3
View tableView in ArticleTable 1. Performance metrics of dehazing results of various dehazing algorithms corresponding to figure 3
Methods PSNR/dB SSIM M1 M2 M3 M4 M1 M2 M3 M4 Method in Ref.［ 7 ］14.517 5 16.613 0 15.756 7 16.945 8 0.765 9 0.827 6 0.871 8 0.775 3 Method in Ref.［ 12 ］16.795 4 19.794 6 15.004 7 22.353 8 0.835 0 0.918 2 0.854 0 0.916 7 Method in Ref.［ 13 ］25.992 7 25.702 2 23.137 1 21.912 7 0.942 4 0.934 0 0.967 1 0.937 9 Method in Ref.［ 14 ］17.620 8 22.227 1 11.287 1 23.464 8 0.845 4 0.894 6 0.403 6 0.936 1 Method in Ref.［ 28 ］17.736 0 16.575 7 17.406 9 17.976 7 0.879 2 0.855 4 0.879 6 0.873 6 Proposed algorithm 36.036 8 30.315 7 27.435 4 30.385 3 0.991 4 0.958 2 0.975 2 0.963 8
Table 2. Average performance metrics of dehazing results with different dehazing algorithms on the SOTS dataset
View tableView in ArticleTable 2. Average performance metrics of dehazing results with different dehazing algorithms on the SOTS dataset
Methods PSNR/dB SSIM Method in Ref.［ 7 ］15.997 2 0.798 8 Method in Ref.［ 12 ］19.147 4 0.878 7 Method in Ref.［ 13 ］22.488 6 0.861 7 Method in Ref.［ 14 ］19.602 4 0.830 1 Method in Ref.［ 28 ］14.919 5 0.759 2 Proposed algorithm 30.214 0 0.945 3
Table 3. Performance metrics of dehazing results of various dehazing algorithms corresponding to Figure 4
View tableView in ArticleTable 3. Performance metrics of dehazing results of various dehazing algorithms corresponding to Figure 4
Methods PSNR/dB SSIM M1 M2 M1 M2 Method in Ref.［ 7 ］14.156 5 13.070 0 0.284 1 0.420 0 Method in Ref.［ 12 ］14.831 3 14.235 9 0.401 7 0.494 3 Method in Ref.［ 13 ］20.477 3 16.966 2 0.516 0 0.533 9 Method in Ref.［ 14 ］14.540 4 15.045 3 0.538 5 0.485 0 Method in Ref.［ 28 ］16.607 8 16.227 2 0.444 4 0.505 8 Proposed algorithm 21.278 1 18.641 5 0.643 4 0.619 9
Table 4. Average performance metrics of dehazing results with different dehazing algorithms on the I-HAZE dataset
View tableView in ArticleTable 4. Average performance metrics of dehazing results with different dehazing algorithms on the I-HAZE dataset
Methods PSNR/dB SSIM Method in Ref.［ 7 ］13.225 0 0.512 1 Method in Ref.［ 12 ］14.765 6 0.561 6 Method in Ref.［ 13 ］15.253 9 0.562 7 Method in Ref.［ 14 ］15.008 8 0.566 4 Method in Ref.［ 28 ］14.735 6 0.484 3 Proposed algorithm 15.472 6 0.591 5
Table 5. Average performance metrics of images restored by different underwater image restoration algorithms on the UIEB dataset
View tableView in ArticleTable 5. Average performance metrics of images restored by different underwater image restoration algorithms on the UIEB dataset
Methods PSNR/dB SSIM Method in Ref.［ 7 ］13.121 3 0.509 0 Method in Ref.［ 35 ］18.746 1 0.816 2 Method in Ref.［ 36 ］17.790 5 0.621 0 Method in Ref.［ 37 ］18.975 7 0.728 2 Proposed algorithm 21.525 3 0.816 5
Tools
Get Citation
Copy Citation Text
Nannan ZHANG, Zhiwei LI, Xinjun GUO, Xinjie XIAO, Hao RUAN. Two-stage image restoration using improved atmospheric scattering model[J]. Optics and Precision Engineering, 2022, 30(18): 2267
Paper Information
Category: Information Sciences
Received: Apr. 21, 2022
Accepted: --
Published Online: Oct. 27, 2022
The Author Email: LI Zhiwei (zhiwei.li@sues.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20