• Laser & Optoelectronics Progress
  • Vol. 59, Issue 10, 1010004 (2022)
Kui Wang and Fuzhen Huang*
Author Affiliations
  • School of Automation Engineering, Shanghai University of Electric Power, Shanghai 200090, China
  • show less
    DOI: 10.3788/LOP202259.1010004 Cite this Article Set citation alerts
    Kui Wang, Fuzhen Huang. Multiscale Retinex Image Enhancement in HSV Space Based on Illumination Compensation[J]. Laser & Optoelectronics Progress, 2022, 59(10): 1010004 Copy Citation Text show less
    Flowchart of proposed algorithm
    Fig. 1. Flowchart of proposed algorithm
    Illumination estimation results of intrinsic protrusion layer. (a) Intrinsic protrusion layer; (b) GKF; (c) BKF; (d) IBKF
    Fig. 2. Illumination estimation results of intrinsic protrusion layer. (a) Intrinsic protrusion layer; (b) GKF; (c) BKF; (d) IBKF
    Reflection layer obtained by intrinsic protrusion layer. (a) Intrinsic protrusion layer; (b) GKF; (c) BKF; (d) IBKF
    Fig. 3. Reflection layer obtained by intrinsic protrusion layer. (a) Intrinsic protrusion layer; (b) GKF; (c) BKF; (d) IBKF
    Bilateral Gamma curves
    Fig. 4. Bilateral Gamma curves
    Illumination compensation layer and brightness transition layer obtained by different φ values.
    Fig. 5. Illumination compensation layer and brightness transition layer obtained by different φ values.
    Gray histograms corresponding to different brightness transition layers and illumination compensation layer. (a) Illumination compensation layer; (b) φ=0.3; (c) φ=0.5; (d) φ=0.8
    Fig. 6. Gray histograms corresponding to different brightness transition layers and illumination compensation layer. (a) Illumination compensation layer; (b) φ=0.3; (c) φ=0.5; (d) φ=0.8
    Comparison of convergence curves before and after sparrow search improvement. (a) Before disturbance; (b) after disturbance
    Fig. 7. Comparison of convergence curves before and after sparrow search improvement. (a) Before disturbance; (b) after disturbance
    Illumination compensation of reflection layer. (a) Brightness transition layer; (b) illumination layer; (c) reflection layer; (d) illumination compensation result
    Fig. 8. Illumination compensation of reflection layer. (a) Brightness transition layer; (b) illumination layer; (c) reflection layer; (d) illumination compensation result
    Comparison of results before and after illumination compensation. (a) Original image; (b) RES1; (c) RES2; (d) proposed algorithm
    Fig. 9. Comparison of results before and after illumination compensation. (a) Original image; (b) RES1; (c) RES2; (d) proposed algorithm
    Comparison of results before and after adaptation. (a) RES3; (b) RES4; (c) RES5; (d) proposed algorithm
    Fig. 10. Comparison of results before and after adaptation. (a) RES3; (b) RES4; (c) RES5; (d) proposed algorithm
    Low illumination image enhancement results under natural light source obtained by different algorithms. (a) Original images; (b) SSR; (c) MSR; (d) MSRCR; (e) method in Ref. [11]; (f) method in Ref. [12];(g) method in Ref. [13]; (h) proposed algorithm
    Fig. 11. Low illumination image enhancement results under natural light source obtained by different algorithms. (a) Original images; (b) SSR; (c) MSR; (d) MSRCR; (e) method in Ref. [11]; (f) method in Ref. [12];(g) method in Ref. [13]; (h) proposed algorithm
    Low illumination image enhancement results under artificial light source obtained by different algorithms. (a) Original images; (b) SSR; (c) MSR; (d) MSRCR; (e) method in Ref. [11]; (f) method in Ref. [12]; (g) method in Ref. [13]; (h) proposed algorithm
    Fig. 12. Low illumination image enhancement results under artificial light source obtained by different algorithms. (a) Original images; (b) SSR; (c) MSR; (d) MSRCR; (e) method in Ref. [11]; (f) method in Ref. [12]; (g) method in Ref. [13]; (h) proposed algorithm
    Evaluation indicatorIntrinsic protrusion layerReflection layer
    GKFBKFIBKF
    Average gradient10.26318.882210.636310.6792
    Table 1. Quality assessment results of reflection layer
    Evaluation indicatorOriginal imageRES1RES2RES3RES4RES5Proposed algorithm
    SD61.238648.311158.634649.067253.781260.783264.2896
    AG10.021311.334911.45439.022910.086011.598812.3413
    IE7.14557.51917.59287.19107.44797.71717.8007
    Table 2. Quality assessment of synthetic results
    SceneEvaluation indicatorOriginal imageSSRMSRMSRCRMethod in Ref.[11Method in Ref.[12Method in Ref.[13Proposed algorithm
    Scene 1SD43.221938.435338.346441.960648.215846.943644.826447.7904
    AG3.75854.80054.79194.94224.65365.27974.67675.6144
    IE6.98676.76716.80037.17947.04277.38407.27497.4987
    Scene 2SD27.807735.571935.508636.408933.939543.170737.632946.1994
    AG3.77325.93865.93115.99305.02175.58534.93856.0015
    IE6.62206.74876.77896.99966.95737.34567.14737.4829
    Scene 3SD32.827335.435735.408737.062938.885847.348538.452346.9623
    AG5.26835.82625.82235.80156.47847.58166.03787.4200
    IE6.88497.03447.04517.14927.14637.50937.25007.5374
    Table 3. Quality evaluation of low illumination image enhancement results under natural light source by different algorithms
    SceneEvaluation indicatorOriginal imageSSRMSRMSRCRMethod in Ref.[11Method in Ref.[12Method in Ref.[13Proposed algorithm
    Scene 4SD43.846749.619749.204449.084350.882151.913847.903253.7664
    AG7.48149.22879.16229.05839.20549.80708.91369.9847
    IE6.99217.13407.28127.52127.24487.58727.41957.6999
    Scene 5SD26.834835.809435.867731.534830.200534.329131.166340.5358
    AG3.16874.52544.52474.56384.11274.44833.74705.2203
    IE6.25886.89776.95036.96106.43376.83706.81467.1465
    Scene 6SD57.364854.255054.171056.455858.460159.200755.086559.1924
    AG4.92236.19066.18456.21476.14056.40105.82526.2239
    IE7.23986.99727.09817.58467.48447.73467.60537.7727
    Table 4. Quality evaluation of low illumination image enhancement results under artificial light source by different algorithms
    Kui Wang, Fuzhen Huang. Multiscale Retinex Image Enhancement in HSV Space Based on Illumination Compensation[J]. Laser & Optoelectronics Progress, 2022, 59(10): 1010004
    Download Citation