Fast Automatic Optic Disc Localization in Retinal Images

Author

Qinghui Cao ; Jianli Liu ; Qiuhong Zhao

Author_Institution

Coll. of Comput. Sci., Beijing Univ. of Technol., Beijing, China

fYear

2013

fDate

26-28 July 2013

Firstpage

827

Lastpage

831

Abstract

Optic disc (OD) localization is an essential step of developing computer-assisted diagnostics. In this paper, we present a fast and robust method to locate the OD center in digital retinal images. The horizontal and vertical location of the OD are be detected respectively, using two main features of the OD, vascular direction and intensity variance. Four public datasets, including 89 normal images and 251 diseased images, were used to evaluate the performance of the proposed method. The success rate is 96.6% with an average computing time of 1.3 s for each image.

Keywords

eye; medical image processing; OD localization; automatic optic disc localization; average computing time; computer-assisted diagnostics; digital retinal images; diseased images; horizontal OD location detection; intensity variance; normal images; performance evaluation; public datasets; success rate; vascular direction; vertical OD location detection; Biomedical optical imaging; Image color analysis; Integrated optics; Optical imaging; Retina; Ultrafast optics; biomedical image processing; optic disc localization; retinal images;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Graphics (ICIG), 2013 Seventh International Conference on

Conference_Location

Qingdao

Type

conf

DOI

10.1109/ICIG.2013.166

Filename

6643785