• Laser & Optoelectronics Progress
  • Vol. 59, Issue 2, 0210013 (2022)
Gengsheng Li, Guojun Liu*, and Wentao Ma
Author Affiliations
  • School of Mathematical Statistics, Ningxia University, Yinchuan , Ningxia 750021, China
  • show less
    DOI: 10.3788/LOP202259.0210013 Cite this Article Set citation alerts
    Gengsheng Li, Guojun Liu, Wentao Ma. Adaptive Image Segmentation Based on Region Information Coupling[J]. Laser & Optoelectronics Progress, 2022, 59(2): 0210013 Copy Citation Text show less
    Original image and preprocessed image. (a) Original image; (b) prepropcessed image
    Fig. 1. Original image and preprocessed image. (a) Original image; (b) prepropcessed image
    Change curve of the weight function. (a) Weight function curve; (b) influence of parameter p on the weight function; (c) influence of parameter k on the weight function
    Fig. 2. Change curve of the weight function. (a) Weight function curve; (b) influence of parameter p on the weight function; (c) influence of parameter k on the weight function
    Segmentation results of our model. (a) Blood vessel image; (b) synthetic image
    Fig. 3. Segmentation results of our model. (a) Blood vessel image; (b) synthetic image
    Segmentation results of different initial contours. (a) Initial contour; (b) segmentation result of 10 iterations; (c) segmentation result of 20 iterations
    Fig. 4. Segmentation results of different initial contours. (a) Initial contour; (b) segmentation result of 10 iterations; (c) segmentation result of 20 iterations
    Segmentation result of our model on the synthetic image. (a) Image 1; (b) image 2; (c) image 3
    Fig. 5. Segmentation result of our model on the synthetic image. (a) Image 1; (b) image 2; (c) image 3
    Segmentation results of blood vessel images with different Gaussian noise. (a) Noisy image; (b) segmentation result
    Fig. 6. Segmentation results of blood vessel images with different Gaussian noise. (a) Noisy image; (b) segmentation result
    Segmentation results of different models. (a) Initial contour; (b) LBF model; (c) model in Ref. [21]; (d) our model
    Fig. 7. Segmentation results of different models. (a) Initial contour; (b) LBF model; (c) model in Ref. [21]; (d) our model
    Segmentation results of natural images. (a) Original image; (b) segmentation result
    Fig. 8. Segmentation results of natural images. (a) Original image; (b) segmentation result
    Image IDJSDSC
    Fig. 3(a1)0.99750.9988
    Fig. 3(a2)0.99770.9989
    Fig. 6(b1)0.98980.9949
    Fig. 6(b2)0.98760.9937
    Fig. 6(b3)0.98680.9933
    Fig. 6(b4)0.98830.9941
    Table 1. JS and DSC values of different images
    Image IDImage size(pixel×pixel)DCSJSPrecision
    Fig. 8(b1)384×2560.94930.90360.9719
    Fig. 8(b2)300×4000.98270.96610.9785
    Fig. 8(b3)288×4000.98770.97570.9865
    Table 2. Evaluation of segmentation results of natural images
    Calculation efficiencyFig. 4(a1)Fig. 4(a2)Fig. 4(a3)Fig. 4(a4)Fig. 4(a5)Fig. 4(a6)
    Number of iterations101513202320
    Time /s1.340.871.001.331.341.16
    Table 3. Number of iterations and time of our model
    Gengsheng Li, Guojun Liu, Wentao Ma. Adaptive Image Segmentation Based on Region Information Coupling[J]. Laser & Optoelectronics Progress, 2022, 59(2): 0210013
    Download Citation