UCRNet: Underwater color image restoration via a polarization-guided convolutional neural network

被引:8
|
作者
Hu, Haofeng [1 ,2 ]
Huang, Yizhao [1 ]
Li, Xiaobo [2 ]
Jiang, Liubing [3 ]
Che, Li [3 ]
Liu, Tiegen [1 ]
Zhai, Jingsheng [2 ]
机构
[1] Tianjin Univ, Sch Precis Instrument & Optoelect Engn, Key Lab Optoelect Informat Technol, Minist Educ, Tianjin, Peoples R China
[2] Tianjin Univ, Sch Marine Sci & Technol, Tianjin, Peoples R China
[3] Guilin Univ Elect Technol, Sch Informat & Commun, Guilin, Peoples R China
基金
中国国家自然科学基金;
关键词
Polarization; polarimetric imaging; scattering media; imaging recovery; physical imaging; TARGET DETECTION; VISION;
D O I
10.3389/fmars.2022.1031549
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Underwater images always suffer from low contrast and color distortion due to the wavelength-dependent scattering and absorption effects caused by particles existing in turbid water, especially in high turbidity conditions. Based on the polarization properties of the backscattering light, polarimetric methods can estimate the intensity level of the backscattering and the transmittance of the media. Accordingly, they can separate the target signal from the undesired ones to achieve high-quality imaging. In addition, learning-based polarimetric methods are effective for gray-model image restoration, but the learning-based polarimetric technique for color image restoration has yet to be considered. In this paper, we propose a 3- dimensional convolutional neural network, which maintains the correlation of polarization information among different polarization channel images as well as embodies polarization constraints, for underwater color image restoration. The experimental results verify that the proposed solution improves the image quality (i.e., the image contrast, details, and color) and outperforms other existing methods, especially when the turbidity of scattering media is high. The proposed solution can be readily applied to practical applications and potentially realize the clear vision in other scattering media, including biomedical imaging and remote sensing.
引用
收藏
页数:11
相关论文
共 50 条
  • [1] Underwater Image Restoration Based on a Parallel Convolutional Neural Network
    Wang, Keyan
    Hu, Yan
    Chen, Jun
    Wu, Xianyun
    Zhao, Xi
    Li, Yunsong
    REMOTE SENSING, 2019, 11 (13)
  • [2] A hybrid attention network with convolutional neural network and transformer for underwater image restoration
    Jiao, Zhan
    Wang, Ruizi
    Zhang, Xiangyi
    Fu, Bo
    Thanh, Dang Ngoc Hoang
    PEERJ COMPUTER SCIENCE, 2023, 9
  • [3] A hybrid attention network with convolutional neural network and transformer for underwater image restoration
    Jiao Z.
    Wang R.
    Zhang X.
    Fu B.
    Thanh D.N.H.
    PeerJ Computer Science, 2023, 9
  • [4] Underwater image restoration via spatially adaptive polarization imaging and color correction
    Li, Yafeng
    Zhang, Jiqing
    Chen, Yuehan
    Li, Yudong
    Tang, Haoming
    Fu, Xianping
    KNOWLEDGE-BASED SYSTEMS, 2024, 305
  • [5] Color restoration of mural image based on double constrained convolutional neural network
    Xu Z.
    Yin W.
    Zhu X.
    Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition), 2020, 48 (06): : 6 - 12
  • [6] Color polarization demosaicking by a convolutional neural network
    Sun, Yuanyuan
    Zhang, Junchao
    Liang, Rongguang
    OPTICS LETTERS, 2021, 46 (17) : 4338 - 4341
  • [7] Multi-Turbidity Underwater Image Restoration Based on Neural Network and Polarization Imaging
    Gui Xinyuan
    Zhang Ran
    Cheng Haoyuan
    Tian Lianbiao
    Chu Jinkui
    LASER & OPTOELECTRONICS PROGRESS, 2022, 59 (04)
  • [8] HSI-DeNet: Hyperspectral Image Restoration via Convolutional Neural Network
    Chang, Yi
    Yan, Luxin
    Fang, Houzhang
    Zhong, Sheng
    Liao, Wenshan
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2019, 57 (02): : 667 - 682
  • [9] Visible light polarization image desmogging via Cycle Convolutional Neural Network
    Zhang, Jingjing
    Zhang, Xin
    Li, Teng
    Zeng, Yuzhou
    Lv, Gang
    Nian, Fudong
    MULTIMEDIA SYSTEMS, 2022, 28 (01) : 45 - 55
  • [10] Visible light polarization image desmogging via Cycle Convolutional Neural Network
    Jingjing Zhang
    Xin Zhang
    Teng Li
    Yuzhou Zeng
    Gang Lv
    Fudong Nian
    Multimedia Systems, 2022, 28 : 45 - 55