• Laser & Optoelectronics Progress
  • Vol. 56, Issue 3, 031008 (2019)
Chendong Cai1, Guanying Huo1、2、*, Yan Zhou1、2, and Hui Han1
Author Affiliations
  • 1 College of Internet of Things Engineering, Hohai University, Changzhou, Jiangsu 213022, China
  • 2 Changzhou Key Laboratory of Sensor Networks and Environment Sensing, Changzhou, Jiangsu 213022, China
  • show less
    DOI: 10.3788/LOP56.031008 Cite this Article Set citation alerts
    Chendong Cai, Guanying Huo, Yan Zhou, Hui Han. Underwater Image Restoration Method Based on Scene Depth Estimation and White Balance[J]. Laser & Optoelectronics Progress, 2019, 56(3): 031008 Copy Citation Text show less
    Calculation of scene depth. (a) Sobel edge; (b) close operation; (c) initial scene depth estimation; (d) regress analysis; (e) scene depth estimation
    Fig. 1. Calculation of scene depth. (a) Sobel edge; (b) close operation; (c) initial scene depth estimation; (d) regress analysis; (e) scene depth estimation
    Comparisons of transmission maps. (a) Original image; (b) RGB histograms; (c) dark channel prior method; (d) proposed method
    Fig. 2. Comparisons of transmission maps. (a) Original image; (b) RGB histograms; (c) dark channel prior method; (d) proposed method
    Overall flowchart of proposed method
    Fig. 3. Overall flowchart of proposed method
    Comparison of restoration results of various underwater environment. (a) Original images; (b) method from Ref. [3]; (c) method from Ref. [15]; (d) method from Ref. [8]; (e) method from Ref. [9]; (f) proposed method
    Fig. 4. Comparison of restoration results of various underwater environment. (a) Original images; (b) method from Ref. [3]; (c) method from Ref. [15]; (d) method from Ref. [8]; (e) method from Ref. [9]; (f) proposed method
    Failure cases of proposed method. (a) Original images; (c) scene depth estimation; (c) transmission map
    Fig. 5. Failure cases of proposed method. (a) Original images; (c) scene depth estimation; (c) transmission map
    MethodLow quality underwater imageLow illumination underwater image
    AveragegradientInformationEntropyUCIQEUIQMAveragegradientInformationEntropyUCIQEUIQM
    Original4.9996.6870.3953.9133.8035.5420.4784.481
    Ref.[3]6.2636.6210.4724.3394.2065.4680.4874.187
    Ref.[15]7.0666.7170.6074.3584.0074.9880.5314.933
    Ref. [8]5.4797.0910.4814.5534.0245.4620.4604.328
    Ref. [9]7.9996.9990.6474.3104.5844.7630.5683.444
    Proposed9.3667.6080.6234.7139.4617.1250.5744.776
    Table 1. Comparisons of average value of objective quality evaluation metrics
    Chendong Cai, Guanying Huo, Yan Zhou, Hui Han. Underwater Image Restoration Method Based on Scene Depth Estimation and White Balance[J]. Laser & Optoelectronics Progress, 2019, 56(3): 031008
    Download Citation