• Laser & Optoelectronics Progress
  • Vol. 58, Issue 4, 0410004 (2021)
Jing Peng, Fengjin Xue*, and Yubin Yuan
Author Affiliations
  • School of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou, Gansu 730070, China
  • show less
    DOI: 10.3788/LOP202158.0410004 Cite this Article Set citation alerts
    Jing Peng, Fengjin Xue, Yubin Yuan. Adaptive Image Defogging Algorithm Combining Multi-Scale Retinex and Dark Channel[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0410004 Copy Citation Text show less
    Flowchart of proposed algorithm
    Fig. 1. Flowchart of proposed algorithm
    Transmittance images after processing by different methods. (a) Hazy images; (b) Ref. [8]; (c) rough estimations of transmittance; (d) transmittance optimization
    Fig. 2. Transmittance images after processing by different methods. (a) Hazy images; (b) Ref. [8]; (c) rough estimations of transmittance; (d) transmittance optimization
    Effect of different algorithms for processing close-range images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    Fig. 3. Effect of different algorithms for processing close-range images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    Effect of different algorithms for processing alternating near-and-far images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    Fig. 4. Effect of different algorithms for processing alternating near-and-far images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    Effect of different algorithms for processing perspective images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    Fig. 5. Effect of different algorithms for processing perspective images. (a) Original images; (b) Ref. [7]; (c) Ref. [8]; (d) Ref. [9]; (e) Ref. [12]; (f) proposed algorithm
    ImageRef. [7]Ref. [8]Ref. [9]Ref. [12]Proposed
    algorithm
    10.09970.08670.03180.01480.1143
    20.24920.29940.10210.17750.4928
    30.34850.3967-0.00210.12800.3821
    40.31150.1456-0.07890.05530.2071
    50.0852-0.0266-0.13920.08490.1729
    60.14510.0823-0.00700.06270.1423
    7-0.0350-0.0150-0.20820.04120.1095
    Mean0.17200.1384-0.04310.08060.2316
    Table 1. p values of processing different images by different algorithms
    ImageRef. [7]Ref. [8]Ref. [9]Ref. [12]Proposed
    algorithm
    11.22961.17881.01591.06571.3386
    21.11381.15961.06851.22691.6675
    31.18591.17611.11691.17511.5083
    41.11041.07740.90771.16791.6777
    51.18141.03410.81251.16251.5929
    61.13711.07501.02291.13971.3475
    71.09391.12300.97991.11591.8649
    Mean1.15031.11770.98921.15051.5711
    Table 2. t values of processing different images by different algorithms
    ImageRef. [7]Ref. [8]Ref. [9]Ref. [12]Proposed algorithm
    161.251361.472362.100365.699362.1086
    255.829857.006958.650158.699563.4518
    358.026158.003361.220464.664861.8904
    459.177455.510060.111865.461763.4308
    560.900559.618461.032567.016262.3975
    660.542459.589961.484668.123863.3756
    759.099260.107061.411866.090460.2116
    Mean59.261058.758360.858865.108062.4095
    Table 3. PSNR of processing different images by different algorithms
    ImageRef. [7]Ref. [8]Ref. [9]Ref. [12]Proposed algorithm
    10.50690.09322.58091.50181.1069
    21.15120.08412.20401.77271.1731
    30.59560.25061.41603.88092.3102
    40.47690.13911.57321.99011.2699
    50.43510.04841.61500.76550.7919
    60.43990.08641.90041.37911.0923
    70.40360.05801.05590.60430.5253
    Mean0.57270.10851.76361.69921.1814
    Table 4. Running time of processing different images by different algorithms unit: s
    Jing Peng, Fengjin Xue, Yubin Yuan. Adaptive Image Defogging Algorithm Combining Multi-Scale Retinex and Dark Channel[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0410004
    Download Citation