Adaptive Thresholding Technique for Retinal Vessel Segmentation Based on GLCM-Energy Information

Table 1

Performance of different segmentation methods on DRIVE database.

Method	Average accuracy	Average sensitivity	Average specificity
Human observer	0.9473	0.7761	0.9725
Staal et al. [16]	0.9442	0.7345	0.9773
Niemeijer et al. [15]	0.9416	0.7145	0.9801
Zana and Klein [8]	0.9377	0.6971	0.9769
Jiang and Mojon [22]	0.9212	0.6399	0.9625
Vlachos and Dermatas [11]	0.9285	0.7468	0.9551
Wang et al. [9]	0.9461	N/A	N/A
Martinez-Perez et al. [7]	0.9181	0.6389	0.9496
Szpak and Tapamo [10]	0.9299	N/A	N/A
Chaudhuri et al. [3]	0.8773	0.3357	0.9794
Soares et al. [17]	0.9466	N/A	N/A
Akram and Khan [20]	0.9469	N/A	N/A
Mendonça and Campilho [12]	0.9463	0.7315	N/A
Marín et al. [19]	0.9452	0.7067	0.9801
Ricci and Perfetti [18]	0.9595	N/A	N/A
Xiao et al. [13]	0.9529	0.7513	0.9792
Yin et al. [14]	0.9267	0.6522	0.9710
Chakraborti et al. [6]	0.9370	0.7205	0.9579
	0.9488	0.7397	0.9691
	0.9511	0.7313	0.9724
	0.9503	0.7375	0.9709
	0.9449	0.7650	0.9623
	0.9477	0.7560	0.9663
	0.9461	0.7632	0.9634