Retinal Vessel Segmentation by Deep Residual Learning with Wide Activation

Table 4

Comparison of the segmentation performance between datasets.

Dataset	Method	Year	Acc	AUC
STARE (trained on DRIVE)	Li et al. [40]	2015	0.9545	0.9671
	Yan et al. [42]	2018	0.9580	0.9678
	DUNet [18]	2018	0.9474	0.9571
	WA-Net	2020	0.9564	0.9707
	Li et al. [40]	2015	0.9486	0.9677
DRIVE (trained on STARE)	Yan et al. [42]	2018	0.9444	0.9568
	DUNet [18]	2018	0.9481	0.9718
	WA-Net	2020	0.9478	0.9700