Optical Instruments, Volume. 46, Issue 5, 9(2024)

Retinal blood vessel segmentation algorithm based on improved U-Net

Yuan LIU... Baicheng LI* and Chunbo WU |Show fewer author(s)
Author Affiliations
  • School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • show less
    Figures & Tables(11)
    Structure of network model
    Residual module
    Detail enhancement model
    Results of U-Net and our algorithm on DRIVE
    Comparison of partial segmentation
    Line chart for comparison of different algorithm indicators
    Comparison of segmentation on DRIVE
    • Table 1. Experimental parameter configuration

      View table
      View in Article

      Table 1. Experimental parameter configuration

      名称参数配置
      优化器Adam
      初始学习率0.001
      训练步数200
      批处理大小4
    • Table 2. Performance indicators

      View table
      View in Article

      Table 2. Performance indicators

      名称缩写公式含义
      敏感性Se$ {\mathrm{Se}}=\dfrac{{\mathrm{TP}}}{{\mathrm{TP}}+{\mathrm{FN}}} $血管被正确分割的指数
      特异性Sp$ {\mathrm{Sp}}=\dfrac{{\mathrm{TN}}}{{\mathrm{TN}}+{\mathrm{FP}}} $图像背景被正确分割的指数
      准确性Acc$ {\mathrm{Acc}}=\dfrac{{\mathrm{TP}}+{\mathrm{TN}}}{{\mathrm{TP}}+{\mathrm{TN}}+{\mathrm{FP}}+{\mathrm{FN}}} $图像整体被正确分割的指数
      F1值F1${\mathrm{ F}}1=\dfrac{{\mathrm{2TP}}}{{\mathrm{2TP}}+{\mathrm{FP}}+{\mathrm{FN}}} $衡量分割结果和标准结果之间相似性的指数
    • Table 3. Results of different structural models based on U-Net

      View table
      View in Article

      Table 3. Results of different structural models based on U-Net

      AlgorithmAccSeSpF1
      U-Net0.96270.81590.98060.8137
      U-Net+Res0.96490.81680.98590.8192
      U-Net+DEA0.96590.83560.98310.8254
      Our algorithm0.96730.83730.98430.8293
    • Table 4. Results of different algorithms on DRIVE

      View table
      View in Article

      Table 4. Results of different algorithms on DRIVE

      MethodAccSeSpF1
      AG-Net[18]0.96540.76380.97390.8132
      CE-Net[19]0.96310.77460.97610.8156
      DUNet[20]0.95790.74390.98720.8265
      SCS-Net[21]0.97680.81570.97210.8241
      CA-Net[22]0.96170.80420.97430.8104
      ASU-Net[23]0.96780.81930.97890.8183
      PVT-CASCADE[24]0.97360.82650.97730.8231
      UNet-2022[25]0.96280.80730.98120.8117
      Our method0.96730.83730.98430.8293
    Tools

    Get Citation

    Copy Citation Text

    Yuan LIU, Baicheng LI, Chunbo WU. Retinal blood vessel segmentation algorithm based on improved U-Net[J]. Optical Instruments, 2024, 46(5): 9

    Download Citation

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

    Category:

    Received: Aug. 28, 2023

    Accepted: --

    Published Online: Jan. 3, 2025

    The Author Email: LI Baicheng (lbcusst@163.com)

    DOI:10.3969/j.issn.1005-5630.202308280111

    Topics