Laser & Optoelectronics Progress, Volume. 57, Issue 14, 141015(2020)

Low-Illumination Remote Sensing Image Enhancement Based on Conditional Generation Adversarial Network

Yanfei Peng**, Tingting Du*, Yi Gao, Lingling Zi, and Yu Sang
Author Affiliations
  • School of Electronic and Information Engineering, Liaoning Technical University, Huludao, Liaoning 125105, China
  • show less
    Figures & Tables(10)
    CGAN model
    Flow chart of proposed algorithm
    Structure of proposed generator
    Structure of proposed discriminator
    Sigmoid function curves
    Examples of synthetic low-illumination remote sensing images
    Comparison of 256 pixel×256 pixel subjective visual images via different algorithms. (a) Original images; (b) HE; (c) MSR; (d) MSRCR; (e) Dong; (f) proposed algorithm
    Results of 380 pixel×380 pixel images processed by proposed algorithm
    Experimental results of different values of α. (a) Original images; (b) α=2.1; (c) α=4; (d) α=6
    • Table 1. Comparison of objective evaluation indexes of different algorithms

      View table

      Table 1. Comparison of objective evaluation indexes of different algorithms

      MethodDongHEMSRMSRCROurs
      PSNR12.67511.21512.50712.76914.710
      MSE3152.29571.99212.78996.2462.82
      SSIM0.21460.19390.26930.23520.3363
      CON1970127355257242528228645
      SAT0.26840.12450.15010.30140.5437
    Tools

    Get Citation

    Copy Citation Text

    Yanfei Peng, Tingting Du, Yi Gao, Lingling Zi, Yu Sang. Low-Illumination Remote Sensing Image Enhancement Based on Conditional Generation Adversarial Network[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141015

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Image Processing

    Received: Oct. 23, 2019

    Accepted: Dec. 11, 2019

    Published Online: Jul. 28, 2020

    The Author Email: Peng Yanfei (pengyf75@126.com), Du Tingting (446154772@qq.com)

    DOI:10.3788/LOP57.141015

    Topics