Table 1.

Algorithms for the localization and segmentation of the OD.

Authors	Method	Database(s) used	No. of images	Performance measure
Lalonde et al. [27]	Pyramidal decomposition, template matching	Non-public dataset	40	ACC 1.00
Lu and Lim [28]	Line operator	DIARETDB0, DIARETDB1, DRIVE, STARE	340	ACC 0.9735
Hoover and Goldbaum [29]	Fuzzy convergence of the retinal vessels	STARE	81	ACC 0.89
Foracchia et al. [31]	Modeling the direction of the retinal vessels	STARE	81	ACC 0.9753
Youssif et al. [32]	2D Gaussian matched filtering, morphological operations	DRIVE; STARE	121	ACC 1.00; ACC 0.9877
Abràmoff and Niemeijer [33]	kNN location regression	Non-public dataset	1000	ACC 0.9990
Sekhar et al. [34]	Morphological operations, Hough transform	DRIVE; STARE	55	ACC 0.947; ACC 0.823
Zhu and Rangayyan [35]	Edge detection, Hough transform	DRIVE	40	ACC 0.9250
Lu [36]	Circular transformation	ARIA, Messidor, STARE	1401	ACC 0.9950
Qureshi et al. [112]	Majority voting-based ensemble	DIARETDB1; DIARETDB1; DRIVE	259	ACC 0.9679; ACC 0.9402; ACC 1.00
Harangi and Hajdu [113]	Weighted majority voting-based ensemble	DIARETDB0; DIARETDB1	219	PPV 0.9846; PPV 0.9887
Hajdu et al. [39]	Spatially constrained majority voting-based ensemble	Non-public dataset; Messidor	1527	ACC 0.921; ACC 0.981
Tomán et al. [40]	Spatially constrained weighted majority voting-based ensemble	Messidor	1200	ACC 0.98
Yu et al. [37]	Template matching, hybrid level-set model	Messidor	1200	ACC 0.9908
Cheng et al. [38]	Superpixel classification	Non-public dataset	650	ACC 0.915