Laser & Optoelectronics Progress, Volume. 59, Issue 16, 1610012(2022)

Information Enhancement Method for Surface Disease Images of Ancient City Walls Based on Adaptive Correction of Illumination Component

Jin Wang1, Huiqin Wang1、*, Ke Wang1, and Zhan Wang2
Author Affiliations
  • 1College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, Shaanxi , China
  • 2Shaanxi Provincial Institute of Cultural Relics Protection, Xi’an 710075, Shaanxi , China
  • show less
    Figures & Tables(15)
    Changes of image brightness before and after correction under different lighting conditions. (a) 2D-Gamma function; (b) enhanced 2D-Gamma function
    Image comparison before and after the enhanced 2D-Gamma function correction. (a) Before the enhanced 2D-Gamma function correction; (b) after the enhanced 2D-Gamma function correction
    Block diagram of homomorphic filtering algorithm
    Homomorphic filter transfer function
    Comparison of brightness histograms of images before and after correction. (a) Brightness histogram before correction; (b) brightness histogram after correction
    Influence of the weight coefficient on the uniformity of illumination
    Image comparison before and after linear weighted fusion. (a) Homomorphic filtered image; (b) linearly weighted fusion image
    Accuracy of edge extraction operator for disease recognition
    Influence of different weighting factors ε on the accuracy of city wall disease recognition
    Flow chart of the proposed method
    Four algorithm processing results. (a) Original images; (b) Gamma function;(c)homomorphic filtering;(d)Enlighten GAN;(e)proposed method
    • Table 1. Image brightness values corrected by enhanced 2D-Gamma function under different lighting conditions

      View table

      Table 1. Image brightness values corrected by enhanced 2D-Gamma function under different lighting conditions

      Image brightness

      vxy

      ixy)=0ixy)=64ixy)=128ixy)=192ixy)=255
      000000
      2013473201.60.004
      40160102405.70.16
      6017712560140.83
      8019014480252.59
      100201161100399.8
      1202111761205618
      1402191901407732
      16022720316010151
      18023321418012573
      200239225200155107
      220245236220189150
      240249247240225205
      255255255255255255
    • Table 2. Comparison of performance indicators of four methods

      View table

      Table 2. Comparison of performance indicators of four methods

      ImageEvaluation indexOriginal imageGamma correctionHomomorphic filteringEnlighten GANProposed method
      Image 1Average illumination5.20065.21269.64166.26206.2450
      Illumination uniformity0.59880.64010.34280.79030.6346
      Image details32.177332.885616.730529.145437.6107
      Average gradient10.2910.435.319.136011.93
      Weighted evaluation12.284312.489010.199811.973414.4327
      Image 2Average illumination5.07875.08968.42575.54206.0403
      Illumination uniformity0.53210.53190.37190.74490.5759
      Image details32.232732.891317.301531.513938.8597
      Average gradient10.1910.335.459.795712.21
      Weighted evaluation12.213412.40229.694012.307114.6647
      Image 3Average illumination5.17285.17468.85026.68446.3356
      Illumination uniformity0.61210.61190.34780.81120.6475
      Image details32.703733.440719.955829.341838.0983
      Average gradient10.4910.656.369.213112.12
      Weighted evaluation12.427512.633510.707712.266414.6300
    • Table 3. Recognition accuracy of four methods

      View table

      Table 3. Recognition accuracy of four methods

      MethodTotal number of test imagesCorrect recognition numberAccuracy /%
      Original image1289674.80
      Gamma correction12810577.95
      Homomorphic filtering12810884.25
      Enlighten GAN1287155.47
      Proposed method12811791.41
    • Table 4. Results of the evaluation of image quality with noise

      View table

      Table 4. Results of the evaluation of image quality with noise

      Noise densityMethodAverage illuminationIllumination uniformityImage detailsAverage gradientWeighted evaluation
      Gamma correction5.51070.606047.893915.023916.9986
      0.02Homomorphic filtering9.03850.423834.243110.591314.9384
      Enlighten GAN5.50750.606147.861215.006216.9876
      Proposed method6.71230.608549.302915.346618.0551
      Gamma correction5.43130.607856.690117.684219.5028
      0.04Homomorphic filtering8.85020.422041.913612.880917.0500
      Enlighten GAN5.42700.608456.631617.688919.4849
      Proposed method6.54050.610257.735917.904620.4016
      Gamma correction8.63940.419047.466714.537218.5375
      0.06Homomorphic filtering5.35980.610363.846219.896821.5387
      Enlighten GAN5.35690.612163.754519.934521.5193
      Proposed method6.36080.612464.712920.057522.3270
      Gamma correction5.29890.613060.846321.768720.9988
      0.08Homomorphic filtering8.53630.407551.230115.654720.3643
      Enlighten GAN5.29940.616369.720120.745623.2147
      Proposed method6.25070.608570.311221.795023.8916
      Gamma correction5.24370.615670.211723.443423.6277
      0.10Homomorphic filtering8.42950.404954.446916.615320.4363
      Enlighten GAN5.24860.620175.051323.553024.7548
      Proposed method6.11660.608875.471323.398425.3772
    Tools

    Get Citation

    Copy Citation Text

    Jin Wang, Huiqin Wang, Ke Wang, Zhan Wang. Information Enhancement Method for Surface Disease Images of Ancient City Walls Based on Adaptive Correction of Illumination Component[J]. Laser & Optoelectronics Progress, 2022, 59(16): 1610012

    Download Citation

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

    Category: Image Processing

    Received: Sep. 14, 2021

    Accepted: Sep. 23, 2021

    Published Online: Jul. 22, 2022

    The Author Email: Wang Huiqin (hqwang@xauat.edu.cn)

    DOI:10.3788/LOP202259.1610012

    Topics