An Automatic Cognitive Graph-Based Segmentation for Detection of Blood Vessels in Retinal Images

Table 4

Illustrative comparison between all possible combinations of inhomogeneity correction and illumination enhancement methods. 1st row: local phase-based method [5]. 2nd row: wavelet-based method [10]. 3rd/7th row: CLAHE correction [3, 4, 41] with multiscale multiorientation BTH/multiscale BTH. 4th/8th row: retinex correction [5] with multiscale multiorientation BTH/multiscale BTH. 5th/9th row: median blurring correction [6] with multiscale multiorientation BTH/multiscale BTH. 6th/10th row: low-pass Gaussian blurring correction with multiscale multiorientation BTH/multiscale BTH.

Method	CII	PSNR
Phase-based	0.412	42.46	0.797
Wavelet-based	0.701	42.76	0.018
CLAHE + multiscale multiorientation BTH	0.528	42.79	0.802
Retinex + multiscale multiorientation BTH	0.475	42.18	0.621
Median + multiscale multiorientation BTH	0.552	43.37	0.818
Gaussian + multiscale multiorientation BTH	0.554	44.18	0.818
CLAHE + multiscale BTH	0.679	44.28	0.816
Retinex + multiscale BTH	0.643	42.71	0.816
Median + multiscale BTH	0.669	44.59	0.818
Gaussian + multiscale BTH	0.714	44.63	0.818