Pith. sign in

REVIEW 3 major objections 5 minor 44 references

LUIEO: A Lightweight Model for Integrating Underwater Image Enhancement and Object Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A lightweight multi-task network claims that underwater image enhancement and object detection improve each other when trained jointly, with detections shown on restored images.

desk verdict A useful lightweight joint detection-enhancement system with an overclaim about being first and an unproven physical-decomposition story on real images. read the letter →

arxiv 2412.07009 v2 pith:HZPIJUL2 submitted 2024-12-01 cs.CV

Bin Li , Li Li , Zhenwei Zhang , Yuping Duan This is my paper · ORCID
classification cs.CV
keywords underwaterimageenhancementobjectdetectionmulti-tasklearninglightweightnetworkphysicalimagingmodelself-supervisedsyntheticsimulationreal-time
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to prove that a single lightweight network can do underwater image enhancement and object detection at the same time, and that the two tasks help each other when trained jointly. Its central claim is that the integration raises detection accuracy—a reported 5.7% mAP50 gain over the detection-only baseline—while also producing visually restored images on which the detections are drawn. To train the enhancement branch without paired clean underwater photos, the method uses a refined physical simulation of underwater imaging as a prior, plus a self-supervision loss that reconstructs the input from predicted clean image, background light, and transmission map. A sympathetic reader would take the paper as showing that multi-task learning with physical constraints is a viable route to deployable real-time underwater perception on limited hardware.

What carries the argument

The load-bearing mechanism is the refined underwater imaging model expressed in formula (3), which extends the standard equation $I_\lambda = J_\lambda t_\lambda + B_\lambda(1-t_\lambda)$ by modeling background light through water depth, attenuation coefficients, and artificial light. The network's physical module predicts $J$, $B$, and $t$; plugging them back into the imaging equation yields a reconstruction loss $L_I = \|I - \tilde I\|_1$ that gives real underwater images a self-supervision signal. The simulation prior from formula (3) is what stops this underdetermined loss from collapsing to trivial decompositions.

What would settle it

Take a held-out set of real underwater images with independently measured water depth and attenuation properties, run the model, and compare its predicted transmission maps and background light to the measured values; if the predictions deviate systematically while detection accuracy stays unchanged, the claimed physical restoration is not what drives the gains.

Watch

Extended reading notes

Core claim

On its own terms, the paper's finding is that jointly optimizing enhancement and detection improves both, provided the enhancement branch is anchored by physics. The network decomposes each underwater image into the three factors of the standard underwater imaging model—clean image $J$, background light $B$, and transmission map $t$—and then recomputes the input image from those factors as a self-supervision signal on real data. Because that reconstruction identity has many solutions, the model is also trained on synthetic images generated by a refined simulation formula that varies water type, depth, and artificial light, which supplies the prior knowledge that makes the decomposition physically meaningful. The experiments report that the integrated model beats a separated enhancement-plus-detection pipeline in both accuracy and speed.

Load-bearing premise

The load-bearing premise is that the hand-designed simulation formula produces synthetic underwater images whose statistics are close enough to real underwater scenes that training on them transfers; if that domain gap is large, the enhancement branch learns an arbitrary decomposition and the reported detection gain would not come from physically correct restoration.

Editorial extensions

If this is right

  • The integrated model runs at 80 FPS with 33.8M parameters, suggesting it fits real-time underwater platforms with limited compute.
  • Joint training raises mAP50 by about 5.7% over the baseline detector, so the enhancement branch is not merely a post-processing step.
  • Detections are drawn on restored images, making it possible to visually verify whether a detection is correct.
  • The model generalizes across color distortion, low light, and blur, thanks to the diverse simulation prior.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the simulation-to-real gap is larger than the paper's experiments suggest, the self-supervision loss could in principle be satisfied by a physically meaningless decomposition, so the detection gain would come from feature changes rather than true restoration; the paper itself concedes this domain gap.
  • The same simulation-plus-physical-constraint recipe could be applied to other underwater tasks such as depth estimation or sonar fusion, where paired labels are also scarce.
  • A stronger test would evaluate the model on real underwater images with independently measured depth and water type, comparing predicted transmission maps and background light against measured values.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript proposes LUIEO, a lightweight multi-task network that simultaneously performs underwater image enhancement and object detection. The enhancement branch decomposes an input image into a clean image, background light, and transmission map via a physical imaging model, and is trained with a self-supervised reconstruction loss on real images plus supervised losses on synthetic images generated by a refined simulation formula. The detection branch is an anchor-free YOLOv8-style head attached to the shared encoder/decoder. The authors report improved UCIQE/UIQM over four enhancement baselines on UIEB, U45, and UCCS, and improved mAP50 over YOLOv8 and other detectors on RUOD and DUO, with 80 FPS and 33.8M parameters. They conclude that joint training and physical constraints improve both sub-tasks.

Significance. The problem is practically relevant: underwater robots need real-time perception and better visualization. The paper's strengths are a clean lightweight design, a comprehensive ablation study (Tables VII and VIII), and public code. The detection results are reproducible in principle, and the ablation suggests simulation prior and physical loss each contribute to mAP. However, the evidence that the model truly enhances real images is limited to no-reference metrics, and the self-supervision on real data is a reconstruction identity that does not by itself enforce a physical decomposition. As a result, the central claim that joint training yields physically correct enhancement and that this enhancement drives the detection gains is not yet established.

major comments (3)
  1. [Section IV-A, Eq. (4)] For real underwater images, Lenhance reduces to LI = ||I - (Jt + B(1-t))||_1. This is a reconstruction identity: any (J,B,t) triple satisfying the imaging model can reproduce I, including degenerate solutions such as t=1,J=I. The only pressure toward a physically meaningful decomposition on real data comes from the simulation prior of Eq. (3) with Table II parameters. Since Section VI concedes a domain gap between synthetic and real images, the physical correctness of the estimated clean image, background light, and transmission on real images is not validated. The paper should test the decomposition directly (e.g., compute PSNR/SSIM on a real paired dataset such as UIEB's reference subset, or compare estimated transmission with known depth) and compare the multi-task gain against a non-physical auxiliary task (e.g., autoencoder reconstruction) to establish that the improvement is due to physical constraints rather than generic regularization.
  2. [Section V-A and Table IV] The quantitative enhancement evaluation relies solely on UCIQE and UIQM, which are no-reference metrics known to be weakly correlated with perceived quality and easily manipulated by color/contrast transforms. The paper notes that UIEB contains 890 paired reference images and that 200 samples were selected for testing, yet no reference-based fidelity metrics (PSNR, SSIM, LPIPS) are reported. Without a paired real-image benchmark or a user study, the claim that 'the proposed model effectively enhances various degraded images' (Section I, Contribution 3) is not supported by the numerical evidence; the visual comparisons in Figs. 5 and 7 are suggestive but not quantitative.
  3. [Section V-F and Table V] The superiority of the integrated model over the separated model LUIEO-S is claimed as evidence that multi-task information exchange benefits both tasks. The differences are small (mAP50 0.755 vs 0.729 on RUOD, 0.695 vs 0.679 on DUO) and no variance, confidence intervals, or significance tests are reported. More importantly, the definition of LUIEO-S is not fully specified: it is described as 'separating the two tasks' but it is unclear whether the enhancement branch is trained without detection gradients, whether the same backbone is used, and whether the detection training set is identical. Without these details, the observed gain could be attributed to training-set augmentation, regularization, or partly to random variation, rather than to a principled information-exchange mechanism.
minor comments (5)
  1. [Section IV-A, Eq. (4)] The symbol cI is used as the weight for the physical loss, but cλ already denotes the attenuation coefficient in Eq. (1); to avoid confusion, rename the weight (e.g., λ_I) in a revision.
  2. [Section V-D, Table III] The choice of cI = 0.5 over cI = 0 is based on a UIQM difference of 4.7000 vs 4.5072 and UCIQE of 0.5084 vs 0.5075 after only 10 epochs on 100 images; the UCIQE difference is negligible, so a more systematic sensitivity analysis or full-training comparison would make the selection more convincing.
  3. [Section V-B] The term 'mAP50-95 c' is not defined consistently; the description says the average over IoU thresholds 0.5, 0.75, and 0.95, whereas the standard mAP50-95 averages over a continuous range from 0.5 to 0.95 with step 0.05. Please clarify.
  4. [Section II-D, Table I] The row labels '1 3 5 7 9' are not explained; it appears they denote additional water types, but this should be stated in the text.
  5. [Fig. 8 caption] The mapping of panels (e)-(h) is inconsistent; the text says '(e-f) are the corresponding prediction results' and 'and (f) is the underwater images calculated by the predicted three variables,' which cannot both be true. Please correct the caption.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the synthetic-prior training is standard supervised inversion of a forward physical model, and the real-image reconstruction loss underdetermination is an identifiability limitation rather than a circular step.

full rationale

The paper's derivation chain is not circular. Synthetic training pairs are generated by the refined imaging formula (3) using NYU-V1 depth maps and the parameter ranges in Table II; the enhancement losses LJ, LB, Lt, and LI compare the network's predicted clean image, background light, transmission map, and reconstructed image to these independently defined synthetic targets. This is standard supervised regression against a forward physical model, not a case where a predicted quantity is defined in terms of the output or where the loss forces the claimed result by construction. For real underwater images, the paper uses only LI = ||I − (Jt + B(1−t))||_1, which is indeed a reconstruction identity admitting infinitely many decompositions, including degenerate ones such as t = 1 and J = I. However, the paper does not claim that LI alone identifies a physical decomposition; it explicitly relies on the simulated prior to regularize the decomposition, and it concedes in Section VI that 'the simulated images cannot be fully approximated to real underwater images, and there is a domain gap between them.' That is an honest identifiability and generalization limitation, not a circular derivation. The object-detection claims are empirical benchmark comparisons (Table V, Table VII, Table VIII) against independently trained baselines and ablations, and the enhancement quality is assessed with no-reference metrics and visual comparisons rather than being derived from the loss. The only overlapping-author citation (UW-CycleGAN, ref. [11]) is used as a comparison method and is not load-bearing for the paper's central claims. Overall, the core contribution is an integrated architecture trained on simulated physics plus real detection labels, and the results are externally testable; any weakness lies in domain transfer and evaluation validity, not in circular reasoning.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities, but it relies on several domain assumptions about the underwater imaging model, the validity of hand-chosen simulation ranges, and the usefulness of no-reference quality metrics. The free parameters are the two loss weights, tuned on evaluation metrics, and the simulation prior ranges, chosen by hand. The self-supervision on real images is a degenerate reconstruction constraint, so the synthetic prior carries much of the load.

free parameters (3)
  • cI = 0.5
    Weight of the physical model loss in Eq. (4), selected by 10-epoch training on 100 random underwater images using UIQM and UCIQE (Section V-D, Table III). Tuned to evaluation metrics.
  • alpha = 0.5
    Weight balancing enhancement and detection losses in Eq. (6), selected by 10-epoch experiments comparing sub-task metrics (Section V-D, Fig. 10). Tuned to evaluation metrics.
  • Simulation prior ranges = See Table II: D 5-20m, c_lambda Table I, ES 0.7-1, Eart 0.7-1, sigma 0.2-1.1, omega_a in [0,1], kappa 0.7-1.1
    Hand-chosen to 'cover as many different underwater environments as possible'; not derived from measured water properties or validated against real underwater statistics.
assumptions (6)
  • domain assumption The underwater imaging model I = J t + B(1-t) (Eq. 1) is an accurate description of the images under study.
    Invoked in Section II-A and used to build the simulation formula, the network decomposition, and the self-supervised loss.
  • domain assumption Background light obeys B = kappa E / c_lambda and illumination E is the sum of surface light and artificial light (Eq. 2).
    Taken from references [24] and [26] and used in Eq. (3) to generate synthetic data and define B^gt in the loss.
  • domain assumption NYU-V1 depth maps can serve as scene distance d in simulated underwater scenes.
    The synthetic dataset is built from NYU-V1 RGB-D images (Section II-D); the depth values are used directly as d(x).
  • domain assumption UCIQE and UIQM are valid proxies for underwater image enhancement quality.
    Used as the only quantitative enhancement metrics (Section V-A, Table IV); both are no-reference and contested in the literature.
  • domain assumption The (J, B, t) decomposition is identifiable from a single image under the proposed training scheme.
    The self-supervised loss LI is a reconstruction identity with many valid decompositions; identifiability is assumed without proof.
  • domain assumption The Jerlov water types in Table I cover the attenuation coefficients of common seawater types.
    Used to sample c_lambda for synthetic image generation (Section II-D, Table I).

how reviews work

0 comments
Cite this review

Pith. "Pith review of LUIEO: A Lightweight Model for Integrating Underwater Image Enhancement and Object Detection." pith.science (2026). https://pith.science/paper/HZPIJUL2

@misc{pith2026241207009,
  author       = {Pith},
  title        = {Pith review of: LUIEO: A Lightweight Model for Integrating Underwater Image Enhancement and Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HZPIJUL2}},
  note         = {Machine review of arXiv:2412.07009}
}
read the original abstract

Underwater optical images inevitably suffer from various degradation factors such as blurring, low contrast, and color distortion, which hinder the accuracy of object detection tasks. Due to the lack of paired underwater/clean images, most research methods adopt a strategy of first enhancing and then detecting, resulting in a lack of feature communication between the two learning tasks. On the other hand, due to the contradiction between the diverse degradation factors of underwater images and the limited number of samples, existing underwater enhancement methods are difficult to effectively enhance degraded images of unknown water bodies, thereby limiting the improvement of object detection accuracy. Therefore, most underwater target detection results are still displayed on degraded images, making it difficult to visually judge the correctness of the detection results. To address the above issues, this paper proposes a multi-task learning method that simultaneously enhances underwater images and improves detection accuracy. Compared with single-task learning, the integrated model allows for the dynamic adjustment of information communication and sharing between different tasks. Due to the fact that real underwater images can only provide annotated object labels, this paper introduces physical constraints to ensure that object detection tasks do not interfere with image enhancement tasks. Therefore, this article introduces a physical module to decompose underwater images into clean images, background light, and transmission images and uses a physical model to calculate underwater images for self-supervision. Numerical experiments demonstrate that the proposed model achieves satisfactory results in visual performance, object detection accuracy, and detection efficiency compared to state-of-the-art comparative methods.

Figures

Figures reproduced from arXiv: 2412.07009 by the authors.

Figure 1
Figure 1. Fig.1. Therefore, the underwater imaging model [23], [24] can [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. A lightweight model integrates image enhancement and object detection. Here, the inverse residual structures are represented as MV2. The image [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Illustration of residual network structure, inverted residual network structure, and spatial pyramid pooling fast (SPPF) structure. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Illustration of MobileViT architecture combining CNN with Trans [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: The enhanced results of proposed model for common underwater degradation types. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: The enhanced results of different values for hyper-parameters [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: The visual comparison among the compared methods on tested datasets. The underwater images are listed in the first row, rows 2-4 show the results [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Results of estimated clean images, transmission maps and background lights, where (a) simulated underwater images, (b) true clean images, (c) the [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 9
Figure 9. Figure 9: Comparison results between the proposed object detection model and the comparison methods on typical underwater degraded images. [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Analysis results of hyper-parameter selection [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: This figure provides a visual comparison of the object detection results between the proposed LUIEO model and the comparative models across [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: Visual comparison of individual SPPF and mobileViT on model [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 31 canonical work pages

  1. [14]

    A lightweight object detection framework for underwater imagery with joint image restoration and color transformation,

    M. Zhou, B. Li, J. Wang, and K. Fu, “A lightweight object detection framework for underwater imagery with joint image restoration and color transformation,” Journal of King Saud University-Computer and Information Sciences , vol. 35, no. 9, p. 101749, 2023

    work page 2023

  2. [18]

    A dual-branch joint learning network for underwater object detection,

    B. Wang, Z. Wang, W. Guo, and Y . Wang, “A dual-branch joint learning network for underwater object detection,” Knowledge-Based Systems , vol. 293, p. 111672, 2024

    work page 2024

  3. [1]

    Cbla: Color balanced locally adjustable underwater image enhancement,

    M. Jha and A. K. Bhandari, “Cbla: Color balanced locally adjustable underwater image enhancement,” IEEE Transactions on Instrumentation and Measurement, 2024

    work page 2024

  4. [2]

    Underwater scene prior inspired deep underwater image and video enhancement,

    C. Li, S. Anwar, and F. Porikli, “Underwater scene prior inspired deep underwater image and video enhancement,” Pattern Recognition, vol. 98, p. 107038, 2020

    2020

  5. [3]

    An underwater image enhancement benchmark dataset and beyond,

    C. Li, C. Guo, W. Ren, R. Cong, J. Hou, S. Kwong, and D. Tao, “An underwater image enhancement benchmark dataset and beyond,” IEEE Transactions on Image Processing , vol. 29, pp. 4376–4389, 2019

    work page 2019

  6. [4]

    Domain adversarial learning towards underwater image enhancement,

    M. Kapoor, R. Baghel, B. N. Subudhi, V . Jakhetiya, and A. Bansal, “Domain adversarial learning towards underwater image enhancement,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 2241–2251

    work page 2023

  7. [5]

    U-shape transformer for underwater image enhancement,

    L. Peng, C. Zhu, and L. Bian, “U-shape transformer for underwater image enhancement,” IEEE Transactions on Image Processing , vol. 32, pp. 3066–3079, 2023

    2023

  8. [6]

    UVEB: A large-scale benchmark and baseline towards real-world underwater video enhancement,

    Y . Xie, L. Kong, K. Chen, Z. Zheng, X. Yu, Z. Yu, and B. Zheng, “UVEB: A large-scale benchmark and baseline towards real-world underwater video enhancement,” in Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition , 2024, pp. 22 358– 22 367

    work page 2024

Show all 44 references
  1. [7]

    Fast underwater image enhancement for improved visual perception,

    M. J. Islam, Y . Xia, and J. Sattar, “Fast underwater image enhancement for improved visual perception,” IEEE Robotics and Automation Letters , vol. 5, no. 2, pp. 3227–3234, 2020

    2020

  2. [8]

    Self-supervised underwater image generation for underwater domain pre-training,

    Z. Wu, Z. Wu, X. Chen, Y . Lu, and J. Yu, “Self-supervised underwater image generation for underwater domain pre-training,” IEEE Transac- tions on Instrumentation and Measurement , 2024

    2024

  3. [9]

    A systematic review and analysis of deep learning-based underwater object detection,

    S. Xu, M. Zhang, W. Song, H. Mei, Q. He, and A. Liotta, “A systematic review and analysis of deep learning-based underwater object detection,” Neurocomputing, vol. 527, pp. 204–232, 2023

    2023

  4. [10]

    Is underwater image enhancement all object detectors need?

    Y . Wang, J. Guo, W. He, H. Gao, H. Yue, Z. Zhang, and C. Li, “Is underwater image enhancement all object detectors need?” IEEE Journal of Oceanic Engineering , vol. 49, no. 2, pp. 606–621, 2024

    2024

  5. [11]

    UW- CycleGAN: Model-driven cyclegan for underwater image restoration,

    H. Yan, Z. Zhang, J. Xu, T. Wang, P. An, A. Wang, and Y . Duan, “UW- CycleGAN: Model-driven cyclegan for underwater image restoration,” IEEE Transactions on Geoscience and Remote Sensing , 2023

    2023

  6. [12]

    Investigating intrinsic degradation factors by multi-branch aggregation for real-world underwa- ter image enhancement,

    X. Xue, Z. Li, L. Ma, Q. Jia, R. Liu, and X. Fan, “Investigating intrinsic degradation factors by multi-branch aggregation for real-world underwa- ter image enhancement,” Pattern recognition, vol. 133, p. 109041, 2023

    2023

  7. [13]

    CURE-Net: a cascaded deep network for underwater image enhancement,

    X. Cai, N. Jiang, W. Chen, J. Hu, and T. Zhao, “CURE-Net: a cascaded deep network for underwater image enhancement,” IEEE journal of oceanic engineering , vol. 49, no. 1, pp. 226–236, 2023

    2023

  8. [15]

    Unitmodule: A lightweight joint image enhancement module for underwater object detection,

    Z. Liu, B. Wang, Y . Li, J. He, and Y . Li, “Unitmodule: A lightweight joint image enhancement module for underwater object detection,” Pattern Recognition, vol. 151, p. 110435, 2024. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 14

    2024

  9. [16]

    Yolotrashcan: A deep learning marine debris detection network,

    W. Zhou, F. Zheng, G. Yin, Y . Pang, and J. Yi, “Yolotrashcan: A deep learning marine debris detection network,” IEEE Transactions on Instrumentation and Measurement , vol. 72, pp. 1–12, 2022

    2022

  10. [17]

    Underwater object detection algorithm based on feature enhancement and progressive dynamic aggregation strategy,

    X. Hua, X. Cui, X. Xu, S. Qiu, Y . Liang, X. Bao, and Z. Li, “Underwater object detection algorithm based on feature enhancement and progressive dynamic aggregation strategy,” Pattern Recognition, vol. 139, p. 109511, 2023

    2023

  11. [19]

    A reinforcement learning paradigm of configuring visual enhancement for object detection in underwater scenes,

    H. Wang, S. Sun, X. Bai, J. Wang, and P. Ren, “A reinforcement learning paradigm of configuring visual enhancement for object detection in underwater scenes,” IEEE Journal of Oceanic Engineering , vol. 48, no. 2, pp. 443–461, 2023

    2023

  12. [20]

    Mobilevitv3: Mobile-friendly vision transformer with simple and effective fusion of local, global and input features,

    S. N. Wadekar and A. Chaurasia, “Mobilevitv3: Mobile-friendly vision transformer with simple and effective fusion of local, global and input features,” arXiv preprint arXiv:2209.15159 , 2022

    2022 arXiv

  13. [21]

    Refined defect detector with deformable transformer and pyramid feature fusion for pcb detection,

    T. Liu, G.-Z. Cao, Z. He, and S. Xie, “Refined defect detector with deformable transformer and pyramid feature fusion for pcb detection,” IEEE Transactions on Instrumentation and Measurement , 2023

    2023

  14. [22]

    Ad-yolo: A real- time yolo network with swin transformer and attention mechanism for airport scene detection,

    W. Zhou, C. Cai, C. Li, H. Xu, and H. Shi, “Ad-yolo: A real- time yolo network with swin transformer and attention mechanism for airport scene detection,” IEEE Transactions on Instrumentation and Measurement, 2024

    2024

  15. [23]

    Visibility in bad weather from a single image,

    R. T. Tan, “Visibility in bad weather from a single image,” in 2008 IEEE conference on computer vision and pattern recognition . IEEE, 2008, pp. 1–8

    2008

  16. [24]

    Underwater image enhancement by wavelength compensation and dehazing,

    J. Y . Chiang and Y .-C. Chen, “Underwater image enhancement by wavelength compensation and dehazing,” IEEE transactions on image processing, vol. 21, no. 4, pp. 1756–1769, 2011

    2011

  17. [25]

    ultralytics/yolov5: v6. 2-yolov5 classification models, apple m1, reproducibility, clearml and deci. ai integrations,

    G. Jocher, A. Chaurasia, A. Stoken, J. Borovec, Y . Kwon, K. Michael, J. Fang, C. Wong, Z. Yifu, D. Montes et al. , “ultralytics/yolov5: v6. 2-yolov5 classification models, apple m1, reproducibility, clearml and deci. ai integrations,” Zenodo, 2022

    2022

  18. [26]

    Deriving inherent optical properties from background color and underwater image enhancement,

    X. Zhao, T. Jin, and S. Qu, “Deriving inherent optical properties from background color and underwater image enhancement,” Ocean Engineering, vol. 94, pp. 163–172, 2015

    2015

  19. [27]

    Indoor scene segmentation using a structured light sensor,

    N. Silberman and R. Fergus, “Indoor scene segmentation using a structured light sensor,” in 2011 IEEE International Conference on Computer Vision Workshops (ICCV workshops) . IEEE, 2011, pp. 601– 608

    2011

  20. [28]

    Light and water: Radiative transfer in natural waters,

    J. R. V . Zaneveld, “Light and water: Radiative transfer in natural waters,” 1995

    1995

  21. [29]

    Revisiting single image depth estimation: Toward higher resolution maps with accurate object boundaries,

    J. Hu, M. Ozay, Y . Zhang, and T. Okatani, “Revisiting single image depth estimation: Toward higher resolution maps with accurate object boundaries,” in 2019 IEEE winter conference on applications of com- puter vision (WACV) . IEEE, 2019, pp. 1043–1051

    2019

  22. [30]

    An efficient advanced- yolov8 framework for thz object detection,

    C. Li, C. Chen, Y . Hei, J. Mou, and W. Li, “An efficient advanced- yolov8 framework for thz object detection,” IEEE Transactions on Instrumentation and Measurement , 2024

    2024

  23. [31]

    Sea-thru: A method for removing water from underwater images,

    D. Akkaynak and T. Treibitz, “Sea-thru: A method for removing water from underwater images,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 1682–1691

    2019

  24. [32]

    A fusion adversarial underwater im- age enhancement network with a public test dataset,

    H. Li, J. Li, and W. Wang, “A fusion adversarial underwater im- age enhancement network with a public test dataset,” arXiv preprint arXiv:1906.06819, 2019

    1906 arXiv

  25. [33]

    A dataset to evaluate underwater image restoration methods,

    A. Duarte, F. Codevilla, J. D. O. Gaya, and S. S. Botelho, “A dataset to evaluate underwater image restoration methods,” in OCEANS 2016- Shanghai. IEEE, 2016, pp. 1–6

    2016

  26. [34]

    Underwater image enhancement via medium transmission-guided multi-color space embedding,

    C. Li, S. Anwar, J. Hou, R. Cong, C. Guo, and W. Ren, “Underwater image enhancement via medium transmission-guided multi-color space embedding,” IEEE Transactions on Image Processing, vol. 30, pp. 4985– 5000, 2021

    2021

  27. [35]

    Twin adversarial contrastive learning for underwater image enhancement and beyond,

    R. Liu, Z. Jiang, S. Yang, and X. Fan, “Twin adversarial contrastive learning for underwater image enhancement and beyond,” IEEE Trans- actions on Image Processing , vol. 31, pp. 4922–4936, 2022

    2022

  28. [36]

    Domain adaptation for underwater image enhancement,

    Z. Wang, L. Shen, M. Xu, M. Yu, K. Wang, and Y . Lin, “Domain adaptation for underwater image enhancement,” IEEE Transactions on Image Processing, vol. 32, pp. 1442–1457, 2023

    2023

  29. [37]

    An underwater color image quality evaluation metric,

    M. Yang and A. Sowmya, “An underwater color image quality evaluation metric,” IEEE Transactions on Image Processing , vol. 24, no. 12, pp. 6062–6071, 2015

    2015

  30. [38]

    Human-visual-system-inspired underwater image quality measures,

    K. Panetta, C. Gao, and S. Agaian, “Human-visual-system-inspired underwater image quality measures,” IEEE Journal of Oceanic Engi- neering, vol. 41, no. 3, pp. 541–551, 2015

    2015

  31. [39]

    Rethinking general underwater object detection: Datasets, challenges, and solutions,

    C. Fu, R. Liu, X. Fan, P. Chen, H. Fu, W. Yuan, M. Zhu, and Z. Luo, “Rethinking general underwater object detection: Datasets, challenges, and solutions,” Neurocomputing, vol. 517, pp. 243–256, 2023

    2023

  32. [40]

    A dataset and benchmark of underwater object detection for robot picking,

    C. Liu, H. Li, S. Wang, M. Zhu, D. Wang, X. Fan, and Z. Wang, “A dataset and benchmark of underwater object detection for robot picking,” in 2021 IEEE international conference on multimedia & expo workshops (ICMEW). IEEE, 2021, pp. 1–6

    2021

  33. [41]

    A gated cross-domain collaborative network for underwater object detection,

    L. Dai, H. Liu, P. Song, and M. Liu, “A gated cross-domain collaborative network for underwater object detection,” Pattern Recognition, vol. 149, p. 110222, 2024

    2024

  34. [42]

    Learning heavily- degraded prior for underwater object detection,

    C. Fu, X. Fan, J. Xiao, W. Yuan, R. Liu, and Z. Luo, “Learning heavily- degraded prior for underwater object detection,” IEEE Transactions on Circuits and Systems for Video Technology , vol. 33, no. 11, pp. 6887– 6896, 2023. Bin Li received the B.S. degree in school of mathe- ...

    2023

  35. [2012]

    Her research interests include numerical optimization, computer vision, and image processing

    She is currently a Professor with the School of Mathematical Sciences, Beijing Normal Univer- sity, Beijing, China. Her research interests include numerical optimization, computer vision, and image processing

  36. [2023]

    His research interests include computer vi- sion and image processing

    He is currently working with the school of Mathematics, North University of China, Taiyuan, China. His research interests include computer vi- sion and image processing. Li Li received the Ph.D. degree from North Uni- versity of China, Taiyuan, China, in 2013. She is currently...

    2013

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.