• Acta Optica Sinica
  • Vol. 40, Issue 2, 0210002 (2020)
Yanbing Xu, Yang Zhou, Canbiao Li, Chujun Zheng*, Rungu Zhang, and Wenbin Wang
Author Affiliations
  • School of Physics and Telecommunication Engineering, South China Normal University, Guangzhou, Guangdong 510006, China
  • show less
    DOI: 10.3788/AOS202040.0210002 Cite this Article Set citation alerts
    Yanbing Xu, Yang Zhou, Canbiao Li, Chujun Zheng, Rungu Zhang, Wenbin Wang. Retinal Vessel Segmentation Based on Super-Pixel Affinity Propagation Clustering[J]. Acta Optica Sinica, 2020, 40(2): 0210002 Copy Citation Text show less
    Framework diagramfor ASLICAP method
    Fig. 1. Framework diagramfor ASLICAP method
    Image pre-processing. (a) Color fundus image; (b) green channel fundus image; (c) CLAHE fundus enhanced image
    Fig. 2. Image pre-processing. (a) Color fundus image; (b) green channel fundus image; (c) CLAHE fundus enhanced image
    B-COSFIRE filter configuration. (a) B-COSFIRE schematic; (b) symmetric B-COSFIRE; (c) asymmetric B-COSFIRE
    Fig. 3. B-COSFIRE filter configuration. (a) B-COSFIRE schematic; (b) symmetric B-COSFIRE; (c) asymmetric B-COSFIRE
    Response mapfor each feature. (a) Hessian maximum eigenvalue; (b) Gabor wavelet transform; (c) B-COSFIRE filter
    Fig. 4. Response mapfor each feature. (a) Hessian maximum eigenvalue; (b) Gabor wavelet transform; (c) B-COSFIRE filter
    Pixel point classification diagram. (a) Initial nearest neighbor classification; (b) KNN reclassification
    Fig. 5. Pixel point classification diagram. (a) Initial nearest neighbor classification; (b) KNN reclassification
    Segmentation diagrams of ASLICAP method in two databases. (a) DRIVE database; (b) STARE database
    Fig. 6. Segmentation diagrams of ASLICAP method in two databases. (a) DRIVE database; (b) STARE database
    Segmentation diagrams of three clustering methods under the same conditions. (a) Original picture; (b) gold standard; (c) ASLICAP; (d) K-means; (e) FCM
    Fig. 7. Segmentation diagrams of three clustering methods under the same conditions. (a) Original picture; (b) gold standard; (c) ASLICAP; (d) K-means; (e) FCM
    DifferenceDRIVE database (K=2500)STARE database (K=4750)
    AccSeSpAccSeSp
    Average0.94630.78790.97250.94300.79300.9581
    Worst0.93580.73520.97220.92030.65770.9413
    Best0.95930.89290.97110.95060.91740.9551
    Table 1. ASLICAP segmentation performance indicators
    AlgorithmDRIVE databaseSTARE database
    AccSeSpAccSeSp
    K-means0.94670.75150.97850.93310.81840.9447
    FCM0.94570.69730.98520.94240.76190.9609
    ASLICAP0.94630.78790.97250.94300.79300.9581
    Table 2. Comparison of performance parameters of ALISCAP method, K-means, and FCM algorithm
    NumMethodDRIVE databaseSTARE database
    AccSeSpAccSeSp
    1Ref. [7]0.9380.7810.9660.8870.7670.939
    2Ref. [13]0.9440.7400.9780.9500.7720.970
    3Ref.[17]0.9340.7250.9660.9410.7510.957
    4Ref. [18]0.9370.7030.9710.9320.7580.950
    5Ref. [19]0.9330.7390.9550.9200.8250.944
    6Ref. [20]0.9380.5690.9930.9460.6380.982
    7Ref. [21]0.9470.7800.9720.9450.7690.938
    8Ref. [22]0.9400.7250.9790.9330.8540.984
    9Proposed method0.9460.7880.9730.9430.7930.958
    Table 3. Comparison of segmentation results of different algorithms
    Yanbing Xu, Yang Zhou, Canbiao Li, Chujun Zheng, Rungu Zhang, Wenbin Wang. Retinal Vessel Segmentation Based on Super-Pixel Affinity Propagation Clustering[J]. Acta Optica Sinica, 2020, 40(2): 0210002
    Download Citation