Optics and Precision Engineering, Volume. 33, Issue 3, 476(2025)
Fast dehazing for large format oblique images based on improved dark channel prior
Figures & Tables(11)
Fig. 7. Dehazing results of different algorithms in light haze scenes
Fig. 8. Dehazing results of different algorithms in dense haze scenes
Table 1. Evaluation metrics of different algorithms in light haze scenes
View tableView in ArticleTable 1. Evaluation metrics of different algorithms in light haze scenes
Method Entropy Average
gradient
PSNR SSIM Raw image 1.773 6 6.864 6 - - PSD［ 13 ］1.939 6 10.342 3 16.845 2 0.513 8 AOD-Net［ 15 ］1.781 8 8.635 7 17.160 3 0.701 4 DCP［ 8 ］1.796 0 8.503 9 17.341 8 0.857 2 Proposed algorithm 1.910 4 8.863 2 17.657 5 0.857 8
Table 2. Evaluation metrics of different algorithms in dense haze scenes
View tableView in ArticleTable 2. Evaluation metrics of different algorithms in dense haze scenes
Method Entropy Average
gradient
PSNR SSIM Raw image 1.631 6 6.375 7 - - PSD［ 13 ］2.142 1 23.870 2 12.987 2 0.587 4 AOD-Net［ 15 ］1.976 4 8.679 3 13.465 4 0.661 6 DCP［ 8 ］1.722 5 7.200 9 12.766 2 0.945 2 Proposed algorithm 2.002 1 8.287 2 13.647 5 0.789 9
Table 3. Processing time of different algorithms
View tableView in ArticleTable 3. Processing time of different algorithms
Method quantity DCP algorithm［ 8 ］Proposed algorithm 1 90.2 s 15.4 s 1 000 24 h 47 m 2 h 11 m
Tools
Get Citation
Copy Citation Text
Huang HUANG, Huan OUYANG, Yang DONG, Xin HE, Xinghao ZHAO. Fast dehazing for large format oblique images based on improved dark channel prior[J]. Optics and Precision Engineering, 2025, 33(3): 476
Paper Information
Category:
Received: Aug. 13, 2024
Accepted: --
Published Online: Apr. 30, 2025
The Author Email: Huan OUYANG (ouyanghuan@whu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20