Table 6. Performance of different segmentation methods, in terms of sensitivity (Se), specificity (Sp), accuracy (Acc) area under the curve (AUC), on the DRIVE and STARE datasets.

Method	DRIVE				STARE
	Se	Sp	Acc	AUC	Se	Sp	Acc	AUC
Second observer	0.776	0.972	0.947	0.874	0.895	0.938	0.934	0.917
Supervised methods
Staal et.al [7]	-	-	0.946	-	-	-	0.951	-
Soares et.al [8]*	-	-	0.946	-	-	-	0.948	-
Lupascu et.al [10]	0.720	-	0.959	-	-	-	-	-
You et.al [13]*	0.741	0.975	0.943	0.858	0.726	0.975	0.949	0.851
Marin et.al [11]	0.706	0.980	0.945	0.843	0.694	0.981	0.952	0.838
Wang et.al [12]*	-	-	0.946	-	-	-	0.952	-
Unsupervised methods
Mendonca et.al [15]*	0.734	0.976	0.945	0.855	0.699	0.973	0.944	0.836
Palomera-Perez et.al [20]	0.660	0.961	0.922	0.811	0.779	0.940	0.924	0.860
Matinez-Perez et.al [19]	0.724	0.965	0.934	0.845	0.750	0.956	0.941	0.853
Al-Diri et.al [16]	0.728	0.955	-	0.842	0.752	0.968	-	0.860
Fraz et.al [14]*	0.715	0.976	0.943	0.846	0.731	0.968	0.944	0.850
Nguyen et.al [21]	-	-	0.940	-	-	-	0.932	-
Bankhead et.al [17]	0.703	0.971	0.9371	0.837	0.758	0.950	0.932	0.854
Orlando et.al [23]	0.785	0.967	-	-	-	-	-	-
Azzopardi et.al [22]	0.766	0.970	0.944	0.961	0.772	0.970	0.950	0.956
Proposed method*	0.744	0.978	0.953	0.861	0.786	0.975	0.951	0.881