Polarimetric binocular three-dimensional imaging in turbid water with multi-feature self-supervised learning

Linghao Shen; Liping Zhang; Pengfei Qi; Xun Zhang; Xiaobo Li; Yizhao Huang; Yongqiang Zhao; Haofeng Hu

doi:10.1186/s43074-025-00185-4

Article Contents

Article Navigation > PhotoniX > > Accepted Manuscript

Linghao Shen, Liping Zhang, Pengfei Qi, Xun Zhang, Xiaobo Li, Yizhao Huang, Yongqiang Zhao, Haofeng Hu. Polarimetric binocular three-dimensional imaging in turbid water with multi-feature self-supervised learning[J]. PhotoniX. doi: 10.1186/s43074-025-00185-4

Citation:

PDF( 4916 KB)

Polarimetric binocular three-dimensional imaging in turbid water with multi-feature self-supervised learning

doi: 10.1186/s43074-025-00185-4

1.
School of Marine Science and Technology, Tianjin University, Tianjin, 300072, China
2.
Martinos Center, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129, USA
3.
School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, 639798, Singapore
4.
School of Automation, Northwestern Polytechnical University, Xi’an, 710072, China

Funds: National Key Research and Development Program of China (2023YFC3108500); National Natural Science Foundation of China (62475190); Tianjin Municipal Science and Technology Bureau (23YFZCSN00230); Special Project for Enterprise Research and Development in Tiankai Higher Education Innovation Park (23YFZXYC00012).

Received Date: 2025-04-03
Accepted Date: 2025-08-06
Rev Recd Date: 2025-07-12

Available Online: 2025-08-21

Abstract

Abstract

Polarization imaging provides significant advantages in underwater environments. However, existing polarization underwater imaging methods primarily focus on leveraging polarization information to suppress the scattering effect to achieve the clear vision, while neglecting other valuable information contained in polarization images, such as the scene depth and the polarization characteristics of the objects. This paper proposes a self-supervised three-dimensional underwater imaging method based on a polarization binocular imager. In addition to improving image quality in turbid water based on polarization imaging, the proposed method merges features from both the enhanced binocular images recovered from polarization information and the feature-rich degree of polarization images into the self-supervised framework to estimate disparities of the scene, achieving high-quality reconstruction of underwater scene depth. We then design multiple self-supervised losses that effectively integrate depth information obtained from both binocular imaging and polarization imaging to guide the learning process. Meanwhile, the proposed method can recover the polarization information of the objects in turbid water, thus enhancing the perception of target properties such as the materials of the objects. Both the simulated experiment and the real-world experiments in the sea demonstrate the effectiveness and superiority of the proposed method.
- Engineering",
- Underwater imaging')">" data-track="click" data-track-action="view keyword" data-track-label="link">Underwater imaging,
- Engineering",
- Polarimetric binocular imaging')">" data-track="click" data-track-action="view keyword" data-track-label="link">Polarimetric binocular imaging,
- Engineering",
- Depth estimation')">" data-track="click" data-track-action="view keyword" data-track-label="link">Depth estimation,
- Engineering",
- Self-supervised learning')">" data-track="click" data-track-action="view keyword" data-track-label="link">Self-supervised learning

FullText(HTML)

References(40)

References

[1]	Zhe SUNXL. Water-related optical imaging: from algorithm to hardware. Sci China Technol Sci. 2025;68(1):1100401.
[2]	Li X, Han Y, Wang H, Liu T, Chen SC, Hu H. Polarimetric imaging through scattering media: a review. Front Phys. 2022;10: 815296.
[3]	Zhu Z, Li X, Zhai J, Hu H. Podb: a learning-based polarimetric object detection benchmark for road scenes in adverse weather conditions. Inf Fusion. 2024;108: 102385.
[4]	Zhu Z, Li X, Ma Q, Zhai J, Hu H. Fdnet: Fourier transform guided dual-channel underwater image enhancement diffusion network. Sci China Technol Sci. 2025;68(1):1100403.
[5]	Rowe MP, Pugh EN, Tyo JS, Engheta N. Polarization-difference imaging: a biologically inspired technique for observation through scattering media. Opt Lett. 1995;20(6):608–10.
[6]	Schechner YY, Karpel N. Recovery of underwater visibility and structure by polarization analysis. IEEE J Oceanic Eng. 2005;30(3):570–87.
[7]	Treibitz T, Schechner YY. Active polarization descattering. IEEE Trans Pattern Anal Mach Intell. 2008;31(3):385–99.
[8]	Powell SB, Garnett R, Marshall J, Rizk C, Gruev V. Bioinspired polarization vision enables underwater geolocalization. Sci Adv. 2018;4(4): eaao6841.
[9]	Li Y, Chen Y, Zhang J, Li Y, Tang H, Fu X. A robust underwater polarization image recovery based on angle of polarization with low-rank and sparse decomposition. Opt Laser Technol. 2025;181: 111669.
[10]	Li X, Hu H, Zhao L, Wang H, Yu Y, Wu L, et al. Polarimetric image recovery method combining histogram stretching for underwater imaging. Sci Rep. 2018;8(1):12430.
[11]	Liu F, Han P, Wei Y, Yang K, Huang S, Li X, et al. Deeply seeing through highly turbid water by active polarization imaging. Opt Lett. 2018;43(20):4903–6.
[12]	Shen L, Reda M, Zhang X, Zhao Y, Kong SG. Polarization-driven solution for mitigating scattering and uneven illumination in underwater imagery. IEEE Trans Geosci Remote Sens. 2024;62:1-15. https://doi.org/10.1109/TGRS.2024.3358828.
[13]	Li X, Xu J, Zhang L, Hu H, Chen SC. Underwater image restoration via Stokes decomposition. Opt Lett. 2022;47(11):2854–7.
[14]	Hu H, Zhang Y, Li X, Lin Y, Cheng Z, Liu T. Polarimetric underwater image recovery via deep learning. Opt Lasers Eng. 2020;133: 106152.
[15]	Lin B, Chen W, Fan X, Peng P, Guo Z. Transformer-based improved U-net for high-performance underwater polarization imaging. Opt Laser Technol. 2025;181: 111664.
[16]	Gao J, Wang G, Chen Y, Wang X, Li Y, Chew KH, et al. Mueller transform matrix neural network for underwater polarimetric dehazing imaging. Opt Express. 2023;31(17):27213–22.
[17]	Hu H, Huang Y, Li X, Jiang L, Che L, Liu T, et al. UCRnet: underwater color image restoration via a polarization-guided convolutional neural network. Front Mar Sci. 2022;9: 1031549.
[18]	Shen L, Xia H, Zhang X, Zhao Y, Li N, Kong SG, et al. U 2 PNet: An Unsupervised Underwater Image-Restoration Network Using Polarization. IEEE Trans Cybern. 2024; 54(9):5164–77. https://ieeexplore.ieee.org/abstract/document/10454091/.
[19]	Guo E, Jiang J, Shi Y, Bai L, Han J. Unsupervised underwater imaging based on polarization and binocular depth estimation. Opt Express. 2024;32(6):9904–19.
[20]	Qi P, Li X, Han Y, Zhang L, Xu J, Cheng Z, et al. U2r-pGAN: unpaired underwater-image recovery with polarimetric generative adversarial network. Opt Lasers Eng. 2022;157: 107112.
[21]	Bai X, Liang Z, Zhu Z, Schwing A, Forsyth D, Gruev V. Polarization-based underwater geolocalization with deep learning. Elight. 2023;3(1):15.
[22]	Jin X, Du D, Jin J, Fan Y. Time-of-flight based imaging in strong scattering underwater environments. Opt Express. 2024;32(21):37247–59.
[23]	Li X, Goudail F, Hu H, Han Q, Cheng Z, Liu T. Optimal ellipsometric parameter measurement strategies based on four intensity measurements in presence of additive Gaussian and Poisson noise. Opt Express. 2018;26(26):34529–46.
[24]	Liu Y, Ding Y, Liu Z, Li X, Tian S, Fan L, et al. Ultrafast laser one-step construction of 3D micro-/nanostructures achieving high-performance zinc metal anodes. PhotoniX. 2024;5(1):6.
[25]	Chen W, Feng S, Yin W, Li Y, Qian J, Chen Q, et al. Deep-learning-enabled temporally super-resolved multiplexed fringe projection profilometry: high-speed kHz 3D imaging with low-speed camera. PhotoniX. 2024;5(1):25.
[26]	Zhang L, Zhan H, Liu X, Cao H, Xing F, You Z. A planar compound eye based microsystem for high precision 3D perception. PhotoniX. 2024;5(1):21.
[27]	Levy D, Peleg A, Pearl N, Rosenbaum D, Akkaynak D, Korman S, et al. Seathru-nerf: Neural radiance fields in scattering media. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. New York: IEEE; 2023. pp. 56–65.
[28]	Jaffe JS. Computer modeling and the design of optimal underwater imaging systems. IEEE J Ocean Eng. 1990;15(2):101–11.
[29]	McGlamery BL. A computer model for underwater camera systems. In: Ocean Optics VI. Bellingham: SPIE; vol. 208. 1980. pp. 221–231.
[30]	Galdran A, Pardo D, Picón A, Alvarez-Gila A. Automatic red-channel underwater image restoration. J Vis Commun Image Represent. 2015;26:132–45.
[31]	Xu Q, Guo Z, Tao Q, Jiao W, Wang X, Qu S, et al. Transmitting characteristics of polarization information under seawater. Appl Opt. 2015;54(21):6584–8.
[32]	He K, Sun J, Tang X. Single image haze removal using dark channel prior. IEEE Trans Pattern Anal Mach Intell. 2010;33(12):2341–53.
[33]	Muja M, Lowe DG. Fast Matching of Binary Features. In: 2012 Ninth Conference on Computer and Robot Vision. New York: IEEE; 2012. pp. 404–410.
[34]	Hirschmuller H. Accurate and efficient stereo processing by semi-global matching and mutual information. In: 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR’05). New York: IEEE; vol. 2. 2005. pp. 807–814.
[35]	Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation. In: Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18. Cham: Springer; 2015. pp. 234–241.
[36]	Zamir SW, Arora A, Khan S, Hayat M, Khan FS, Yang MH, et al. Multi-stage progressive image restoration. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. New York: IEEE; 2021. pp. 14821–14831.
[37]	Kendall A, Gal Y, Cipolla R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. In: Proceedings of the IEEE conference on computer vision and pattern recognition. New York: IEEE; 2018. pp. 7482–7491.
[38]	Xu H, Zhang J, Cai J, Rezatofighi H, Yu F, Tao D, et al. Unifying flow, stereo and depth estimation. IEEE Trans Pattern Anal Mach Intell. 2023;45(11):13941–58. https://doi.org/10.1109/TPAMI.2023.3298645.
[39]	Ranftl R, Bochkovskiy A, Koltun V. Vision transformers for dense prediction. In: Proceedings of the IEEE/CVF international conference on computer vision. New York: IEEE; 2021. pp. 12179–12188.
[40]	Eigen D, Puhrsch C, Fergus R. Depth map prediction from a single image using a multi-scale deep network. Adv Neural Inf Process Syst. 2014;27:2366–74.