Laser & Optoelectronics Progress, Volume. 57, Issue 24, 241505(2020)

Single Image Dehazing Method Based on Self-Adaptive Bilateral-Filtering with Boundary Restriction

Wei Hu1,2 and Yin Gao1、*
Author Affiliations
  • 1Quanzhou Institute of Equipment Manufacturing, Haixi Institute, Chinese Academy of Sciences, Quanzhou, Fujian 362000, China
  • 2College of Electrical and Control Engineering, North University of China, Taiyuan, Shanxi 0 38507, China
  • show less
    Figures & Tables(5)
    Segmentation results of different types of bright regions. (a)(b) Original image and segmentation result of wall; (c)(d) original image and segmentation result of garden; (e)(f) original image and segmentation result of mountain
    Dehazing results of natural images. (a) Original images; (b) He's algorithm; (c) Zhao's algorithm; (d) Hu's algorithm; (e) proposed algorithm
    Dehazing results of synthetic images. (a) Original images; (b) He's algorithm; (c) Zhao's algorithm;(d) Hu's algorithm; (e) proposed algorithm; (f) dehazed images
    • Table 1. Algorithm evaluation for natural images

      View table

      Table 1. Algorithm evaluation for natural images

      Image in figure 2AlgorithmeΣr¯NR-FQA
      Orange forestWithout algorithm processing----0.73
      Algorithm in Ref. [2]0.2602.04-4.65
      Algorithm in Ref. [9]0.0801.932.89
      Algorithm in Ref. [10]0.0801.731.71
      Proposed algorithm0.3301.95-4.09
      GrasslandWithout algorithm processing----5.45
      Algorithm in Ref. [2]0.2701.32-8.98
      Algorithm in Ref. [9]-0.010.0401.36-11.23
      Algorithm in Ref. [10]0.2601.39-16.23
      Proposed algorithm0.2601.47-9.43
      BuildingWithout algorithm processing---10.15
      Algorithm in Ref. [2]1.5401.62-20.70
      Algorithm in Ref. [9]1.1202.48-23.07
      Algorithm in Ref. [10]1.190.0051.77-20.92
      Proposed algorithm1.3001.94-32.02
      ForestWithout algorithm processing---20.67
      Algorithm in Ref. [2]0.8101.6313.74
      Algorithm in Ref. [9]0.5302.542.23
      Algorithm in Ref. [10]0.410.0202.13-9.13
      Proposed algorithm0.7002.00-10.87
    • Table 2. Algorithm evaluation of synthetic images

      View table

      Table 2. Algorithm evaluation of synthetic images

      Image in figure 3AlgorithmeΣr¯SSIM
      RoadWithout algorithm processing----
      Algorithm in Ref. [2]1.010.071.650.66
      Algorithm in Ref. [9]0.510.102.090.84
      Algorithm in Ref. [10]0.410.032.550.84
      Proposed algorithm0.660.092.450.84
      GardenWithout algorithm processing----
      Algorithm in Ref. [2]0.510.021.320.84
      Algorithm in Ref. [9]0.400.081.960.80
      Algorithm in Ref. [10]0.430.201.620.74
      Proposed algorithm0.360.121.810.90
      ChurchWithout algorithm processing----
      Algorithm in Ref. [2]0.3201.530.83
      Algorithm in Ref. [9]0.200.061.860.85
      Algorithm in Ref. [10]0.150.081.830.83
      Proposed algorithm0.170.092.480.91
      GrasslandWithout algorithm processing----
      Algorithm in Ref. [2]0.610.021.850.76
      Algorithm in Ref. [9]0.3102.040.87
      Algorithm in Ref. [10]0.530.122.290.72
      Proposed algorithm0.540.062.550.85
    Tools

    Get Citation

    Copy Citation Text

    Wei Hu, Yin Gao. Single Image Dehazing Method Based on Self-Adaptive Bilateral-Filtering with Boundary Restriction[J]. Laser & Optoelectronics Progress, 2020, 57(24): 241505

    Download Citation

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

    Category: Machine Vision

    Received: Apr. 9, 2020

    Accepted: Jun. 17, 2020

    Published Online: Nov. 18, 2020

    The Author Email: Yin Gao (yngaoyin@163.com)

    DOI:10.3788/LOP57.241505

    Topics