Dehazing Evaluation: Real-World Benchmark Datasets, Criteria, and Baselines

Zhao, Shiyu; Zhang, Lin; Huang, Shuaiyi; Shen, Ying

doi:10.1109/tip.2020.2995264

Cited by 107 publications

(47 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We used the recent real-world outdoor benchmark dataset BeDDE [6] for comparison experiments. BeDDE contains images with different haze levels, as well as relatively clear reference images.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Single Image Dehazing Algorithm Based on Modified Dark Channel Prior

Zhou

Zhang

Liu

et al. 2021

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Single image dehazing algorithm based on Dark Channel Prior (DCP) is widely known. More and more image dehazing algorithms based on DCP have been proposed. However, we found that it is more effective to use DCP in the RAW images before the ISP pipeline. In addition, for the problem of DCP failure in the sky area, we propose an algorithm to segment the sky region and compensate the transmission. Extensive experimental results on both subjective and objective evaluation demonstrate that the performance of the modified DCP (MDCP) has been greatly improved, and it is competitive with the state-of-the-art methods.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Then we compare MDCP with the state-of-the-art methods by visibility index (VI) and realness index (RI) [6] indicators for BeDDE, as shown in Table 1. From these results, we can find that MDCP achieves the best performance for RI indicator.…”

Section: Methodsmentioning

confidence: 99%

Single Image Dehazing Algorithm Based on Modified Dark Channel Prior

Zhou

Zhang

Liu

et al. 2021

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Menurut [2]- [4], kabut dan asap mengarah ke rendahnya penglihatan karena partikel-partikel tersebut akan memecah dan menyebarkan sinar. Salah satu penyebab terjadinya kabut disebabkan adanya partikelpartikel atmosfer yang berkumpul di udara.…”

Section: Pendahuluanunclassified

“…Setelah DM diperoleh dari persamaan (8), maka nilai PT bisa diperoleh dari persamaan (2). Akan tetapi, untuk meningkatkan kualitas t, pada penelitian ini menerapkan metode koreksi gamma.…”

Section: Estimasi Ptunclassified

Perbaikan Visibilitas pada Citra Berkabut Kawah Gunung Berapi Kelud Menggunakan Color Attenuation Prior

Putra

Firdaus

Utama

2021

mib

View full text Add to dashboard Cite

Mount Kelud is one of the volcanoes that erupted in 2014. To observe the activity, CCTV has been installed in the crater peak of Mount Kelud. The crater of Mount Kelud emits gases coming from the bottom of the crater. The gas makes CCTV surveillance undisturbed so that the resulting image will have noise. By using dehazing, visibility can be improved so that the resulting image can be seen clearly. The method used for this operation is Color Attenuation Prior. In the early stages, there is a dark channel process that works to turn low-intensity pixels dark. The second step is to estimate the atmospheric light from the dark channel image. This process is almost in parallel with the estimation of the depth map. The fourth step is estimating the transmission map, which functions to transmit low-intensity pixels to high-intensity pixels. The last one is the radiance recovery scene. The results show that the foggy image is successful, and it eliminates the fog effect, thus enhancing the visibility of the image. And from questionnaire results, we got 80 % positive results from all respondents. Further research so that it can be applied directly or in real-time.

show abstract

“…Other available datasets include: RESIDE [39], D-HAZE [40], I-HAZE [41], O-HAZE [42], HazeRD [43], KITTI [44], and BeDDE [45].…”

Section: Experiments 41 Datasetsmentioning

confidence: 99%

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

2021

KSII TIIS

View full text Add to dashboard Cite

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 performance 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 lightweight 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 network causes problems, such as loss of object details and color in some image areas, the multipatch 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 hazy maps 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 stateof-the-art methods, such as PSNR, SSIM, LPIPS, and individual subjective evaluation.

show abstract

Dehazing Evaluation: Real-World Benchmark Datasets, Criteria, and Baselines

Cited by 107 publications

References 58 publications

Single Image Dehazing Algorithm Based on Modified Dark Channel Prior

Single Image Dehazing Algorithm Based on Modified Dark Channel Prior

Perbaikan Visibilitas pada Citra Berkabut Kawah Gunung Berapi Kelud Menggunakan Color Attenuation Prior

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

Contact Info

Product

Resources

About