Structure-based level set method for automatic retinal vasculature segmentation

EURASIP Journal on Image and Video Processing

Table 3 Statistical results of our method for test images of 1 to 20 from DRIVE dataset

Dataset	Image number	Se	Sp	Ppv	Npv	Acc	κ
DRIVE	1	0.8182	0.9581	0.7461	0.9723	0.9398	0.7457
	2	0.7764	0.9654	0.7982	0.9608	0.9371	0.7502
	3	0.7387	0.9513	0.7218	0.9551	0.9202	0.6834
	4	0.7456	0.9677	0.7826	0.9607	0.9378	0.7279
	5	0.7419	0.9682	0.7878	0.9593	0.9371	0.7279
	6	0.7142	0.9726	0.8126	0.9534	0.9358	0.7233
	7	0.7507	0.9466	0.6846	0.9609	0.9204	0.6699
	8	0.7285	0.9619	0.7332	0.9610	0.9325	0.6923
	9	0.7223	0.9738	0.7880	0.9630	0.9439	0.7221
	10	0.7535	0.9656	0.7507	0.9661	0.9400	0.7179
	11	0.7456	0.9512	0.6972	0.9613	0.9243	0.6768
	12	0.7932	0.9594	0.7391	0.9697	0.9383	0.7297
	13	0.7165	0.9665	0.7811	0.9533	0.9307	0.7073
	14	0.7940	0.9573	0.7130	0.9720	0.9380	0.7160
	15	0.7867	0.9463	0.6321	0.9742	0.9296	0.6616
	16	0.7880	0.9697	0.7989	0.9677	0.9457	0.7621
	17	0.7581	0.9715	0.7900	0.9660	0.9451	0.7425
	18	0.8407	0.9517	0.6965	0.9784	0.9388	0.7271
	19	0.8696	0.9599	0.7504	0.9815	0.9489	0.7764
	20	0.8254	0.9604	0.7162	0.9785	0.9458	0.7364