Opto-Electronic Engineering, Volume. 51, Issue 6, 240063-1(2024)

A multilayer linear fusion algorithm for endoscopic image enhancement

Shuangyuan Wang1,*... Zhiyuan Yao1, Yurong Zhang1, Huaiqi Xue1 and Gengsheng He2 |Show fewer author(s)
Author Affiliations
  • 1School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • 2China Southern Power Grid Energy Development Research Institute Co.,Ltd., Guangzhou, Guangdong 510530, China
  • show less
    Figures & Tables(11)
    Flowchart of the multilayer fusion algorithm
    (a) Origin; (b) Layer1 luminance enhancement algorithm; (c) Layer 2 CLAHE algorithm; (d) Layer 3 linear fusion algorithm
    Comparison of different algorithms for three sets of images (a-c) of Pylorus
    Comparison of different algorithms for three sets of images (a-c) of Retroflex-stomach
    Comparison of different algorithms for three sets of images (a-c) of Z-line
    • Table 1. The contents of the HyperKvasir partial dataset

      View table
      View in Article

      Table 1. The contents of the HyperKvasir partial dataset

      1st layer2nd layer3rd layerNo. of images
      Upper GIAnatomical landmarksPylorus999
      Retroflex-stomach764
      Z-line932
      Pathological findingsBarrett's41
      Barrett's Short Segments53
      Esophagitis Grage-A403
      Esophagitis Grage-B-D260
    • Table 2. NIQE

      View table
      View in Article

      Table 2. NIQE

      图像原始图像CLAHEGCAGCWDRetinex-NetZero-DCEENDOIMLE本文算法
      Pylorus3.7763.3713.7623.5863.6263.8933.7053.543
      Retroflex3.5383.3853.8023.5693.7894.2653.6283.373
      Z-line3.6363.0693.2733.4713.8214.5233.6753.112
    • Table 3. PSNR

      View table
      View in Article

      Table 3. PSNR

      图像CLAHEGCAGCWDRetinex-NetZero-DCEENDOIMLE本文算法
      Pylorus17.188.2313.4517.6116.3714.3419.43
      Retroflex12.999.1513.9118.5117.6115.4118.96
      Z-line13.779.7714.9818.7718.0315.4519.56
    • Table 4. SSIM

      View table
      View in Article

      Table 4. SSIM

      图像CLAHEGCAGCWDRetinex-NetZero-DCEENDOIMLE本文算法
      Pylorus0.9380.6550.8960.7110.7870.8810.955
      Retroflex0.8930.6530.8940.7780.7370.8840.936
      Z-line0.9130.7510.9010.7930.7640.8710.946
    • Table 5. Ablation experiment

      View table
      View in Article

      Table 5. Ablation experiment

      GC (未线性拉伸)自适应GC改进CLAHE本文算法
      PSNR8.5610.2315.6819.55
      SSIM0.5790.7730.8970.948
      NIQE3.7853.3683.1863.246
    • Table 6. Run time

      View table
      View in Article

      Table 6. Run time

      算法CLAHEGCAGCWDRetinex-NetZero-DCEENDOIMLE本文算法
      时间/s40.3512.186.213.450.015.464.56
    Tools

    Get Citation

    Copy Citation Text

    Shuangyuan Wang, Zhiyuan Yao, Yurong Zhang, Huaiqi Xue, Gengsheng He. A multilayer linear fusion algorithm for endoscopic image enhancement[J]. Opto-Electronic Engineering, 2024, 51(6): 240063-1

    Download Citation

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

    Category:

    Received: Mar. 18, 2024

    Accepted: May. 31, 2024

    Published Online: Oct. 21, 2024

    The Author Email: Wang Shuangyuan (王双园)

    DOI:10.12086/oee.2024.240063

    Topics