LEPF-Net: Light Enhancement Pixel Fusion Network for Underwater Image Enhancement

被引:0
|
作者
Yan, Jiaquan [1 ]
Wang, Yijian [2 ]
Fan, Haoyi [3 ]
Huang, Jiayan [4 ]
Grau, Antoni [5 ]
Wang, Chuansheng [5 ]
机构
[1] Minjiang Univ, Coll Comp & Control Engn, Fujian Prov Key Lab Informat Proc & Intelligent Co, Fuzhou 350121, Peoples R China
[2] Minjiang Univ, Coll Math & Data Sci, Fuzhou 350108, Peoples R China
[3] Zhengzhou Univ, Sch Comp & Artificial Intelligence, Zhengzhou 450001, Peoples R China
[4] Putian Univ, New Engn Ind Coll, Putian 351100, Peoples R China
[5] Univ Politecn Cataluna, Dept Automatic Control Tech, Barcelona 08034, Spain
关键词
underwater image enhancement; light enhancement curve; pixel fusion; adaptive feature weights; RESTORATION; WATER;
D O I
10.3390/jmse11061195
中图分类号
U6 [水路运输]; P75 [海洋工程];
学科分类号
0814 ; 081505 ; 0824 ; 082401 ;
摘要
Underwater images often suffer from degradation due to scattering and absorption. With the development of artificial intelligence, fully supervised learning-based models have been widely adopted to solve this problem. However, the enhancement performance is susceptible to the quality of the reference images, which is more pronounced in underwater image enhancement tasks because the ground truths are not available. In this paper, we propose a light-enhanced pixel fusion network (LEPF-Net) to solve this problem. Specifically, we first introduce a novel light enhancement block (LEB) based on the residual block (RB) and the light enhancement curve (LE-Curve) to restore the cast color of the images. The RB is adopted to learn and obtain the feature maps from an original input image, and the LE-Curve is used to renovate the color cast of the learned images. To realize the superb detail of the repaired images, which is superior to the reference images, we develop a pixel fusion subnetwork (PF-SubNet) that adopts a pixel attention mechanism (PAM) to eliminate noise from the underwater image. The PAM adapts weight allocation to different levels of a feature map, which leads to an enhancement in the visibility of severely degraded areas. The experimental results show that the proposed LEPF-Net outperforms most of the existing underwater image enhancement methods. Furthermore, among the five classic no-reference image quality assessment (NRIQA) indicators, the enhanced images obtained by LEPF-Net are of higher quality than the ground truths from the UIEB dataset.
引用
收藏
页数:20
相关论文
共 50 条
  • [21] NUAM-Net: A Novel Underwater Image Enhancement Attention Mechanism Network
    Wen, Zhang
    Zhao, Yikang
    Gao, Feng
    Su, Hao
    Rao, Yuan
    Dong, Junyu
    JOURNAL OF MARINE SCIENCE AND ENGINEERING, 2024, 12 (07)
  • [22] LPS-Net: Lightweight Parallel Strategy Network for Underwater Image Enhancement
    Jiang, Jingxia
    Huang, Peiyun
    Tong, Lihan
    Yin, Junjie
    Chen, Erkang
    APPLIED SCIENCES-BASEL, 2023, 13 (16):
  • [23] IDC-Net: integrated dynamic context network for underwater image enhancement
    Li, Zenglu
    Guo, Xiaoyu
    Wu, Xiaohua
    JOURNAL OF ELECTRONIC IMAGING, 2024, 33 (05)
  • [24] Swin transformer and fusion for underwater image enhancement
    Sun, Jinghao
    Dong, Junyu
    Lv, Qingxuan
    INTERNATIONAL WORKSHOP ON ADVANCED IMAGING TECHNOLOGY (IWAIT) 2022, 2022, 12177
  • [25] Effective Single Underwater Image Enhancement by Fusion
    Fang, Shuai
    Deng, Rong
    Cao, Yang
    Fang, Chunlong
    JOURNAL OF COMPUTERS, 2013, 8 (04) : 904 - 911
  • [26] Underwater Image Enhancement by Wavelet Based Fusion
    Khan, Amjad
    Ali, Syed Saad Azhar
    Malik, Aamir Saeed
    Anwer, Atif
    Meriaudeau, Fabrice
    2016 IEEE 6TH INTERNATIONAL CONFERENCE ON UNDERWATER SYSTEM TECHNOLOGY: THEORY AND APPLICATIONS, 2016, : 83 - 88
  • [27] Color Balance and Fusion for Underwater Image Enhancement
    Ancuti, Codruta O.
    Ancuti, Cosmin
    De Vleeschouwer, Christophe
    Bekaert, Philippe
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2018, 27 (01) : 379 - 393
  • [28] PGN: Progressively Guided Network with Pixel-Wise Attention for Underwater Image Enhancement
    Jia, Huidi
    Wang, Qiang
    Fu, Bo
    Zheng, Zhimin
    Tang, Yandong
    APPLIED SCIENCES-BASEL, 2025, 15 (02):
  • [29] A deep Retinex network for underwater low-light image enhancement
    Ji, Kai
    Lei, Weimin
    Zhang, Wei
    MACHINE VISION AND APPLICATIONS, 2023, 34 (06)
  • [30] A deep Retinex network for underwater low-light image enhancement
    Kai Ji
    Weimin Lei
    Wei Zhang
    Machine Vision and Applications, 2023, 34