The new approach to automatic detection of Optic Disc from non-dilated retinal images

Author

Vahabi, Z. ; Vafadoost, M. ; Gharibzadeh, Shahriar

Author_Institution

Dept. of Biomed. Eng., Amirkabir Univ. of Technol., Tehran, Iran

fYear

2010

fDate

3-4 Nov. 2010

Firstpage

1

Lastpage

6

Abstract

Diabetic retinopathy is the commonest cause of blindness. Diabetes causes cataracts, Glaucoma and diabetic retinopathy. The Optic Disc is the exit point of retinal nerve fibers from the eye and the entrance and exit point for retinal blood vessels. The detection of Optic Disc is very essential to locate the various anatomical features in the retinal images. We describe a new filtering approach in the wavelet domain for image preprocessing. Sobel edge detection, Texture Analysis, Intensity and Template matching was used to detect Optic Disc. The proposed algorithm is tested on 150 images of Messidor dataset. Experimental results indicates that we are able to achieve 87.54% sensitivity, 99.76% specificity and 99.81% accuracy.

Keywords

biomedical optical imaging; blood vessels; diseases; edge detection; eye; feature extraction; image matching; medical image processing; wavelet transforms; Messidor dataset; Sobel edge detection; anatomical features; automatic detection; blindness; cataracts; diabetic retinopathy; filtering approach; glaucoma; image preprocessing; intensity matching; nondilated retinal images; optic disc; retinal blood vessels; retinal nerve fibers; template matching; texture analysis; wavelet domain; Accuracy; Image edge detection; Image segmentation; Lighting; Retina; Diabetic Retinopathy; Entropy; Gradient; Morphlogical Operation; Optic Disk; Template matching; wavelet packet;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Engineering (ICBME), 2010 17th Iranian Conference of

Conference_Location

Isfahan

Print_ISBN

978-1-4244-7483-7

Type

conf

DOI

10.1109/ICBME.2010.5704960

Filename

5704960