Optics and Precision Engineering, Volume. 33, Issue 5, 802(2025)
Adaptive color compensation and multi-scale fusion for underwater image enhancement
Figures & Tables(17)
Fig. 1. Flow chart of underwater image enhancement algorithm based on adaptive color compensation and multi-scale fusion
Fig. 3. Color compensation and white light balance algorithm processing results
Fig. 8. Schematic diagram of NSCT tower-type multi-scale decomposition and multi-directional decomposition
Fig. 9. Schematic diagram of NSDFB Secondary Multi-Directional Decomposition
Table 1. PSNR values of images
View tableView in ArticleTable 1. PSNR values of images
方法 Image 1 Image 2 Image 3 Image 4 Image 5 Image 6 Image 7 Image 8 平均值 原始值 9.446 3 16.624 1 16.042 2 17.375 1 19.453 2 15.023 7 15.411 2 14.446 1 15.477 7 DCP 9.083 7 14.236 2 13.812 0 10.966 4 12.800 7 13.171 7 12.385 6 12.029 9 12.310 8 UCM 15.739 8 16.053 5 16.050 3 15.282 2 20.021 4 19.587 8 14.691 1 13.794 6 16.402 6 AWB 16.743 6 16.902 2 20.052 2 12.376 5 12.421 3 15.760 1 11.575 4 13.503 1 14.916 8 Fu 21.493 8 18.179 1 21.599 1 16.803 1 20.214 8 19.816 2 19.680 5 18.102 5 19.486 1 本文 24.483 1 18.934 5 22.689 1 20.076 5 23.998 0 21.980 1 22.526 0 23.022 1 22.213 7
Table 2. SSIM values of images
View tableView in ArticleTable 2. SSIM values of images
方法 Image 1 Image 2 Image 3 Image 4 Image 5 Image 6 Image 7 Image 8 平均值 原始值 0.032 2 0.642 6 0.578 0 0.740 2 0.663 2 0.524 0 0.567 0 0.520 9 0.533 5 DCP 0.083 0 0.591 8 0.512 0 0.658 5 0/0.477 3 0.456 3 0.506 9 0.537 9 0.478 UCM 0.447 8 0.333 0 0.689 6 0.282 1 0.884 8 0.766 4 0.698 6 0.572 9 0.584 4 AWB 0.592 0 0.412 6 0.743 0 0.450 8 0.516 8 0.665 7 0.499 1 0.538 8 0.552 4 Fu 0.814 1 0.448 4 0.770 6 0.600 9 0.869 6 0.767 0 0.792 7 0.750 4 0.726 7 本文 0.876 0 0.669 2 0.876 7 0.837 7 0.889 2 0.863 0 0.890 9 0.868 5 0.846 4
Table 3. IE values of images
View tableView in ArticleTable 3. IE values of images
方法 Image 1 Image 2 Image 3 Image 4 Image 5 Image 6 Image 7 Image 8 平均值 原始值 5.866 6 7.208 6 6.874 9 6.656 3 7.444 1 7.445 0 7.064 5 6.577 4 6.892 2 DCP 5.207 9 6.791 0 6.919 0 6.634 4 7.001 3 6.857 4 6.957 5 6.784 9 6.644 2 UCM 7.349 8 7.677 5 7.432 4 7.388 1 7.598 3 7.636 3 7.277 4 7.270 3 7.453 8 AWB 7.209 3 7.538 0 7.485 8 6.792 4 7.002 9 7.432 0 6.961 2 6.783 1 7.150 6 Fu 7.664 0 7.708 1 7.661 5 7.784 9 7.662 7 7.653 6 7.740 8 7.692 5 7.696 本文 7.804 6 7.722 0 7.793 6 7.655 6 7.713 1 7.737 4 7.804 6 7.796 8 7.753 5
Table 4. UCIQE values of images
View tableView in ArticleTable 4. UCIQE values of images
方法 Image 1 Image 2 Image 3 Image 4 Image 5 Image 6 Image 7 Image 8 平均值 原始值 0.357 0 0.407 1 0.272 1 0.257 5 0.476 6 0.461 4 0.352 8 0.226 8 0.351 4 DCP 0.383 2 0.469 8 0.400 7 0.343 2 0.585 6 0.515 0 0.419 9 0.347 9 0.433 2 UCM 0.489 6 0.491 6 0.486 8 0.446 2 0.524 5 0.490 9 0.406 0 0.446 7 0.472 8 AWB 0.327 0 0.408 3 0.365 2 0.400 6 0.526 1 0.526 4 0.465 6 0.344 4 0.420 5 Fu 0.440 0 0.438 9 0.396 1 0.479 6 0.434 4 0.483 7 0.442 6 0.437 4 0.444 1 本文 0.446 0 0.455 6 0.430 3 0.434 1 0.503 9 0.507 9 0.476 9 0.459 7 0.464 3
Table 5. Objective evaluation indexes of ablation experiment
View tableView in ArticleTable 5. Objective evaluation indexes of ablation experiment
指标 去掉色彩补偿 去掉白平衡 去掉伽马校正 去掉瑞利拉伸 去掉锐化 本文 PSNR 18.849 4 18.988 0 22.605 2 15.776 2 16.821 9 22.811 7 SSIM 0.799 8 0.710 5 0.866 0 0.635 5 0.719 6 0.869 9 IE 7.617 8 6.613 1 7.740 4 6.433 6 7.015 3 7.741 7 UCIQE 0.474 3 0.443 3 0.446 1 0.312 5 0.340 6 0.453 6
Tools
Get Citation
Copy Citation Text
Cuiyin LIU, Yuanshuai CHEN, Tengyun JING, Feng WANG. Adaptive color compensation and multi-scale fusion for underwater image enhancement[J]. Optics and Precision Engineering, 2025, 33(5): 802
Paper Information
Category:
Received: Oct. 25, 2024
Accepted: --
Published Online: May. 20, 2025
The Author Email: LIU Cuiyin (liucuiyin@kust.edu)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20