• Laser & Optoelectronics Progress
  • Vol. 62, Issue 2, 0237011 (2025)
Yang Tao*, Hao Tan, and Liqun Zhou
Author Affiliations
  • School of Communications and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
  • show less
    DOI: 10.3788/LOP241207 Cite this Article Set citation alerts
    Yang Tao, Hao Tan, Liqun Zhou. ULCF-Net: An Underwater Low-Illumination Image Enhancement Algorithm Based on a Cross-Scale Structure and Color Fusion[J]. Laser & Optoelectronics Progress, 2025, 62(2): 0237011 Copy Citation Text show less
    Structure of ULCF-Net
    Fig. 1. Structure of ULCF-Net
    Structure of BEM
    Fig. 2. Structure of BEM
    Structure of artifact suppression sampling module. (a) Process of coding end sampling; (b) process of inverse operation
    Fig. 3. Structure of artifact suppression sampling module. (a) Process of coding end sampling; (b) process of inverse operation
    Cross-scale connections. (a) Original cross-scale connection; (b) cross-scale connection Ⅰ; (c) cross-scale connection Ⅱ
    Fig. 4. Cross-scale connections. (a) Original cross-scale connection; (b) cross-scale connection Ⅰ; (c) cross-scale connection Ⅱ
    Strcture of DSMCEM
    Fig. 5. Strcture of DSMCEM
    Experimental results of feature points matching
    Fig. 6. Experimental results of feature points matching
    Subjective comparison on test set A between proposed algoritm and low light image enhancement algorithms
    Fig. 7. Subjective comparison on test set A between proposed algoritm and low light image enhancement algorithms
    Subjective comparison on test set A between proposed algoritm and underwater image enhancement algorithms
    Fig. 8. Subjective comparison on test set A between proposed algoritm and underwater image enhancement algorithms
    Subjective comparison results on test set C by different algorithms
    Fig. 9. Subjective comparison results on test set C by different algorithms
    AlgorithmPSNR /dBSSIMMSELPIPSUIQMUCIQE
    LAU-Net2421.46830.86010.00780.12393.15830.5993
    HWMNet1820.28360.83740.01050.15443.17960.5912
    Retinexformer2520.77490.85650.00990.12463.07830.5895
    Enlighten Anything2615.32810.72250.03240.26832.85060.5384
    FourLLIE1921.64780.85370.00850.12733.25990.6063
    CDAN2720.96180.86760.00940.11843.20800.6015
    ULCF-Net (proposed)23.27290.87810.00530.09233.11500.6106
    Table 1. Comparison results on test set A between proposed algoritm and low light image enhancement algorithms
    AlgorithmPSNR /dBSSIMMSELPIPSUIQMUCIQE
    LDS-Net2118.12890.79500.01670.18853.29210.5702
    U-shape2822.50640.77360.00660.16353.21690.5807
    Five A+[1421.88920.85620.00780.12943.16930.5931
    LANet2921.37210.83160.00920.10293.04980.6099
    STSC3018.54560.54830.01640.12053.09350.5995
    Shallow-UWnet3112.89150.64480.06080.21862.87250.2946
    ULCF-Net (proposed)23.27290.87810.00530.09233.11500.6106
    Table 2. Comparison results on test set A between proposed algoritm and underwater image enhancement algorithms
    AlgorithmUIQMUCIQE
    LDS-Net212.78270.5672
    HWMNet182.49270.5898
    Enlightn Anything261.53380.5591
    FourLLIE192.71370.5842
    CDAN272.45730.5758
    LAU-Net242.25480.5830
    STSC302.59920.5861
    Shallow-UWnet311.81270.5335
    ULCF-Net (proposed)2.37030.5916
    Table 3. Objectively evaluate results on test set C by different algorithms
    BEM formCEM typePSNR /dBSSIMMSE
    BEM(a)BEM(b)SKFFDSMCEM
    15.44850.69120.0305
    17.10210.75730.0216
    19.90570.83410.0116
    19.79850.83280.0119
    16.83620.76560.0248
    22.89390.87120.0059
    23.27290.87810.0053
    Table 4. Module to module ablation experiment
    Cross-scale connection formPSNR /dBSSIMMSE
    Origin
    22.87180.86990.0063
    23.15460.87010.0062
    23.27290.87810.0053
    Table 5. Cross-scale connection ablation experiment
    Yang Tao, Hao Tan, Liqun Zhou. ULCF-Net: An Underwater Low-Illumination Image Enhancement Algorithm Based on a Cross-Scale Structure and Color Fusion[J]. Laser & Optoelectronics Progress, 2025, 62(2): 0237011
    Download Citation