Bayesian Method with Spatial Constraint for Retinal Vessel Segmentation

Table 1

Performance of vessel segmentation methods (DRIVE images).

Method	Average accuracy	Sensitivity	Specificity
Staal et al. [2]	0.9442	0.7194	0.9773
You et al. [3]	0.9434	0.7410	0.9751
Marín et al. [4]	0.9452	0.7067	0.9801
Niemeijer et al. [5]	0.9417	0.6898	0.9696
Zhang et al. [11]	0.9382	0.7120	0.9724
Yin et al. [12]	0.9267	0.6252	0.9710
Fraz et al. [15]	0.9430	0.7152	0.9769
Miri and Mahloojifar [16]	0.9458	0.7352	0.9795
Mendonça and Campilho [18]	0.9452	0.7344	0.9764
Martinez-Perez et al. [21]	0.9344	0.7246	0.9655
Martinez-Perez et al. [22]	0.9220	0.6602	0.9612
Vlachos and Dermatas [23]	0.9291	0.7472	0.9550
Espona et al. [29]	0.9352	0.7436	0.9615
Proposed Method	0.9529	0.7513	0.9792