Chinese Journal of Liquid Crystals and Displays, Volume. 40, Issue 8, 1177(2025)

Image defogging based on superpixel segmentation and fusion of dark and light channel

Ning MA1,2, Xia CHANG2,3、*, and Weibing ZHANG2,3
Author Affiliations
  • 1School of Computer Science and Engineering, North Minzu University, Yinchuan 750021, China
  • 2Ningxia Key Laboratory of Intelligent Information and Big Data Processing, Northern Minzu University, Yinchuan 750021, China
  • 3School of Mathematics and Information Science, North Minzu University, Yinchuan 750021, China
  • show less
    Figures & Tables(15)
    Comparison of foggy day images (a) and dark channel images (b)
    Architecture of the proposed method
    Comparison of the results of different filtering windows
    Fogged images and their histograms
    Dark channel image and adaptive thresholding segmentation results
    Comparison of traditional dark channel and mixing channel
    Dark channel fusion results
    Comparison of recovered images before and after window correction
    Comparison of dark area segmentation results
    Comparison of outdoor scene image restoration
    Comparison of indoor scene image restoration
    Histogram of objective evaluation indexes of image defogging results of different algorithms. (a) PSNR; (b) SSIM; (c) Defogging time of different algorithms.
    • Table 1. Comparison of mean PSNR values of image defogging by different methods

      View table
      View in Article

      Table 1. Comparison of mean PSNR values of image defogging by different methods

      算法OTSO-HAZEI-HAZE
      NLD2419.26714.8114.612
      DehazeNet2320.75515.06614.177
      DCP917.29314.86111.975
      BCCR2120.93914.58112.035
      MSCNN2220.47617.17115.018
      CAP2015.49815.39313.468
      MGF2523.64816.94314.997
      SSDC-F26.81518.47117.637
    • Table 2. Comparison of mean values of SSIM for image defogging by different methods

      View table
      View in Article

      Table 2. Comparison of mean values of SSIM for image defogging by different methods

      算法OTSO-HAZEI-HAZE
      NLD240.7720.3880.47
      DehazeNet230.7250.3570.465
      DCP90.8170.3090.382
      BCCR210.8620.2720.435
      MSCNN220.8930.3760.525
      CAP200.5690.3250.417
      MGF250.8750.5170.554
      SSDC-F0.9390.5760.698
    • Table 3. Comparison of defogging time of different methods

      View table
      View in Article

      Table 3. Comparison of defogging time of different methods

      算法OTSO-HAZEI-HAZE
      NLD246.11839.80381.252
      DehazeNet232.48950.83289.475
      DCP96.22560.77670.122
      BCCR212.70122.47141.416
      MSCNN221.44370.55976.151
      CAP201.58216.83151.537
      MGF251.58812.25241.956
      SSDC-F1.52129.87136.281
    Tools

    Get Citation

    Copy Citation Text

    Ning MA, Xia CHANG, Weibing ZHANG. Image defogging based on superpixel segmentation and fusion of dark and light channel[J]. Chinese Journal of Liquid Crystals and Displays, 2025, 40(8): 1177

    Download Citation

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

    Category:

    Received: Mar. 20, 2025

    Accepted: --

    Published Online: Sep. 25, 2025

    The Author Email: Xia CHANG (changxia0104@163.com)

    DOI:10.37188/CJLCD.2025-0061

    Topics