• KSII Transactions on Internet and Information Systems
    Monthly Online Journal (eISSN: 1976-7277)

Lightweight multiple scale-patch dehazing network for real-world hazy image

Vol. 15, No. 12, December 31, 2021
10.3837/tiis.2021.12.009, Download Paper (Free):

Abstract

Image dehazing is an ill-posed problem which is far from being solved. Traditional image dehazing methods often yield mediocre effects and possess substandard processing speed, while modern deep learning methods perform best only in certain datasets. The haze removal effect when processed by said methods is unsatisfactory, meaning the generalization perfor-mance fails to meet the requirements. Concurrently, due to the limited processing speed, most dehazing algorithms cannot be employed in the industry. To alleviate said problems, a light-weight fast dehazing network based on a multiple scale-patch framework (MSP) is proposed in the present paper. Firstly, the multi-scale structure is employed as the backbone network and the multi-patch structure as the supplementary network. Dehazing through a single net-work causes problems, such as loss of object details and color in some image areas, the multi-patch structure was employed for MSP as an information supplement. In the algorithm image processing module, the image is segmented up and down for processed separately. Secondly, MSP generates a clear dehazing effect and significant robustness when targeting real-world homogeneous and nonhomogeneous hazymaps and different datasets. Compared with existing dehazing methods, MSP demonstrated a fast inference speed and the feasibility of real-time processing. The overall size and model parameters of the entire dehazing model are 20.75M and 6.8M, and the processing time for the single image is 0.026s. Experiments on NTIRE 2018 and NTIRE 2020 demonstrate that MSP can achieve superior performance among the state-of-the-art methods, such as PSNR, SSIM, LPIPS, and individual subjective evaluation.


Statistics

Show / Hide Statistics

Statistics (Cumulative Counts from December 1st, 2015)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article

[IEEE Style]
J. Wang, C. Ding, M. Wu, Y. Liu and G. Chen, "Lightweight multiple scale-patch dehazing network for real-world hazy image," KSII Transactions on Internet and Information Systems, vol. 15, no. 12, pp. 4420-4438, 2021. DOI: 10.3837/tiis.2021.12.009.

[ACM Style]
Juan Wang, Chang Ding, Minghu Wu, Yuanyuan Liu, and Guanhai Chen. 2021. Lightweight multiple scale-patch dehazing network for real-world hazy image. KSII Transactions on Internet and Information Systems, 15, 12, (2021), 4420-4438. DOI: 10.3837/tiis.2021.12.009.

[BibTeX Style]
@article{tiis:25146, title="Lightweight multiple scale-patch dehazing network for real-world hazy image", author="Juan Wang and Chang Ding and Minghu Wu and Yuanyuan Liu and Guanhai Chen and ", journal="KSII Transactions on Internet and Information Systems", DOI={10.3837/tiis.2021.12.009}, volume={15}, number={12}, year="2021", month={December}, pages={4420-4438}}