Automatic Detection of Optic Disc in Retinal Image by Using Keypoint Detection, Texture Analysis, and Visual Dictionary Techniques

Table 3

Quantitative results of the various methods.

Authors	Technique	Dataset	Number of images	Accuracy (%)
Azam et al. [16]	Bright region and vessel density	DRIVE	40	100
Lu and Lim [17]	Iterative Savitzky-Golay smoothing procedure, thresholding, and morphologic operations	DIARETDB0; DIARETDB1; DRIVE	259	96.91 (average)
Godse and Bormane [18]	Bright region and density criterion	DIARETDB1; DRIVE	89; 40	96.62; 100
Lupascu et al. [19]	Hough transform	DRIVE	40	95
Sekhar et al. [20]	Hough transform	DRIVE	40	94.4
Qureshi et al. [21]	Pyramidal decomposition, edge detection, entropy filter, and Hough transformation	DIARETDB1; DRIVE	89; 40	94.02; 100
Ying et al. [22]	Fractal method and local histogram	DRIVE	40	97.5
The proposed method	Bright region, keypoint detection, VD, and classification	DIARETDB1; DRIVE; ROC	89; 40; 100	94.38; 95; 90