Laser & Optoelectronics Progress, Volume. 60, Issue 16, 1610004(2023)

Image Threshold Segmentation Method Based on Cumulative Residual Information Energy

Jing Liu*, Yue Tian, and Jiulun Fan
Author Affiliations
  • School of Communication and Information Engineering, Xi'an University of Post & Telecommunications, Xi'an 710121, Shaanxi, China
  • show less
    Figures & Tables(24)
    Probability distribution function
    Cumulative distribution function
    Cumulative residual distribution function
    Cumulative residual information energy function
    Color images of test images. (a) 368016; (b) 135069; (c) 12074; (d) 100007; (e) 238011
    Gray images of test images. (a) 368016; (b) 135069; (c) 12074; (d) 100007; (e) 238011
    Gray histograms of test images. (a) 368016; (b) 135069; (c) 12074; (d) 100007; (e) 238011
    Segmentation results of different methods (#368016). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#135069). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#12074). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#100007). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#238011). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Comparison of segmentation accuracy of test images under different methods
    Comparison of peak signal-to-noise ratio of test images under different methods
    Color images of test images. (a) cell1; (b) cell2; (c) cell3
    Gray histograms of test images. (a) cell1; (b) cell 2; (c) cell 3
    Segmentation results of different methods (#cell1). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#cell 2). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    Segmentation results of different methods (#cell3). (a) Standard segmentation; (b) method 1; (c) method 2; (d) method 3; (e) method 4; (f) method 5; (g) method 6; (h) method 7
    • Table 1. Running time of brute force and recursive algorithm

      View table

      Table 1. Running time of brute force and recursive algorithm

      Image numberBrute force(considering symmetry)Recursive algorithm
      Average value81.14180.1452
      36801690.94960.1471
      13506978.56730.1409
      1207478.08050.1407
      10000780.44770.1333
      23801177.66410.1638
    • Table 2. Comparison of segmentation accuracy of test images under different methods

      View table

      Table 2. Comparison of segmentation accuracy of test images under different methods

      Image numberMethod 1Method 2Method 3Method 4Method 5Method 6Method 7
      Average value76.1946.0875.7368.3459.5984.1687.05
      36801640.6722.8336.4371.1227.3987.5089.16
      13506996.3845.4096.3993.3867.3496.4496.22
      1207476.6745.2775.5445.8565.9876.4476.90
      10000783.6244.3886.8666.4873.8176.5189.06
      23801183.6072.5083.4364.8663.4583.9183.92
    • Table 3. Comparison of peak signal-to-noise ratio of test images under different methods

      View table

      Table 3. Comparison of peak signal-to-noise ratio of test images under different methods

      Image numberMethod 1Method 2Method 3Method 4Method 5Method 6Method 7
      Average value45.472343.843545.429044.656044.071545.697245.8953
      36801644.192743.204743.909844.142843.274545.788445.9916
      13506949.379046.347149.386848.624846.114149.430549.3727
      1207444.571142.952244.503342.932543.699844.548444.5544
      10000746.339044.111846.534945.176845.519045.755746.5860
      23801142.879642.601742.810142.403241.750342.962842.9716
    • Table 4. Comparison of segmentation accuracy of test images under different methods

      View table

      Table 4. Comparison of segmentation accuracy of test images under different methods

      Image numberMethod 1Method 2Method 3Method 4Method 5Method 6Method 7
      Average value92.2548.5891.7051.3925.7179.1894.07
      cell191.8851.5291.4252.3614.0368.7998.67
      cell294.2140.8893.9548.0411.7693.8194.34
      cell390.6553.3489.7353.7851.3574.9489.21
    • Table 5. Comparison of peak signal-to-noise ratio of test images under different methods

      View table

      Table 5. Comparison of peak signal-to-noise ratio of test images under different methods

      Image numberMethod 1Method 2Method 3Method 4Method 5Method 6Method 7
      Average value40.214139.913540.121539.891138.544039.549740.2371
      cell136.556136.809036.638036.860036.174536.829436.8690
      cell235.744935.446735.703235.474034.791234.797935.7911
      cell348.341447.484848.023347.339544.666447.022048.0512
    Tools

    Get Citation

    Copy Citation Text

    Jing Liu, Yue Tian, Jiulun Fan. Image Threshold Segmentation Method Based on Cumulative Residual Information Energy[J]. Laser & Optoelectronics Progress, 2023, 60(16): 1610004

    Download Citation

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

    Category: Image Processing

    Received: Jul. 15, 2022

    Accepted: Oct. 13, 2022

    Published Online: Aug. 15, 2023

    The Author Email: Liu Jing (liujing121777@163.com)

    DOI:10.3788/LOP222085

    Topics