Two-step approach for single underwater image enhancement

Fu, Xueyang; Fan, Zhiwen; Mei, Ling; Huang, Yue; Ding, Xinghao

doi:10.1109/ispacs.2017.8266583

Cited by 191 publications

(142 citation statements)

References 22 publications

Supporting

Mentioning

142

Contrasting

Order By: Relevance

“…For the downward looking images and forward looking images shown in Figs. 7 and 8, the fusion-based [31], retinexbased [33], and histogram prior [45] significantly remove the effect of haze on the underwater images while the twostep-based [32], Red Channel [42], and blurriness-based [46] remain some haze in the results. We note that UDCP [37], regression-based [39], and GDCP [40] tend to bring in color deviation in the enhanced results.…”

Section: A Qualitative Evaluationmentioning

confidence: 98%

“…Subjective comparisons on bluish and greenish underwater images. From left to right are raw underwater images, and the results of fusion-based [31], retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images. , retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images.…”

Section: B Quantitative Evaluationmentioning

confidence: 99%

“…From left to right are raw underwater images, and the results of fusion-based [31], retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images. , retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images. , retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images.…”

Section: B Quantitative Evaluationmentioning

confidence: 99%

“…Subjective comparisons on low backscatter scenes. From left to right are raw underwater images, and the results of fusion-based [31], retinex-based [33], two-step-based [32], UDCP [37], Red Channel [42], histogram prior [45], regression-based [39], blurriness-based [46], GDCP [40], and reference images. Fig.…”

Section: B Quantitative Evaluationmentioning

confidence: 99%

“…The dive+ and fusion-based [31] are no more the best performers. It [31] 0.6414 1.5310 two-step-based [32] 0.5776 1.4002 retinex-based [33] 0.6062 1.4338 UDCP [37] 0.5852 1.6297 regression-based [39] 0.5971 1.2996 GDCP [40] 0.5993 1.4301 Red Channel [42] 0.5421 1.2147 histogram prior [45] 0.6778 1.5440 blurriness-based [46] 0.6001 1.3757 dive+ 0.6227 1.3410 is interesting that the good performers in terms of UCIQE and UIQM metrics are not consistent with the subjective pairwise comparisons, though both UCIQE and UIQM claim that they take the human visual perception into account. Such a result provides evidence that the current image quality evaluation metrics designed for underwater image are inconsistent with human visual perception in some cases.…”

Section: Raws Fusion-based Retinex-based Udcp Red Channel Histogram Pmentioning

confidence: 99%

See 4 more Smart Citations

An Underwater Image Enhancement Benchmark Dataset and Beyond

Guo

Ren

et al. 2020

IEEE Trans. on Image Process.

1,242

672

View full text Add to dashboard Cite

Underwater image enhancement has been attracting much attention due to its significance in marine engineering and aquatic robotics. Numerous underwater image enhancement algorithms have been proposed in the last few years. However, these algorithms are mainly evaluated using either synthetic datasets or few selected real-world images. It is thus unclear how these algorithms would perform on images acquired in the wild and how we could gauge the progress in the field. To bridge this gap, we present the first comprehensive perceptual study and analysis of underwater image enhancement using large-scale real-world images. In this paper, we construct an Underwater Image Enhancement Benchmark (UIEB) including 950 realworld underwater images, 890 of which have the corresponding reference images. We treat the rest 60 underwater images which cannot obtain satisfactory reference images as challenging data. Using this dataset, we conduct a comprehensive study of the stateof-the-art underwater image enhancement algorithms qualitatively and quantitatively. In addition, we propose an underwater image enhancement network (called Water-Net) trained on this benchmark as a baseline, which indicates the generalization of the proposed UIEB for training Convolutional Neural Networks (CNNs). The benchmark evaluations and the proposed Water-Net demonstrate the performance and limitations of state-of-the-art algorithms, which shed light on future research in underwater image enhancement. The dataset and code are available at https://li-chongyi.github.io/proj benchmark.html.

show abstract