A Method for Segmentation of Retinal Image Vessels

Author

Huang, Shuying ; Zhang, Erhu

Author_Institution

Dept. of Inf. Sci., Xi´´an Univ. of Technol.

Volume

2

fYear

0

fDate

0-0 0

Firstpage

9673

Lastpage

9676

Abstract

Retinal image vessels can reflect different states of many diseases, which are often used for the registration and mosaic of retinal images. Therefore, it is very important to accurately extract blood vessels. The paper firstly analyzes the characteristic of retinal images, and secondly gray mathematical morphology theories are used to smooth and strengthen retinal images in order to remove the background and enhance the brightness of retinal blood vessels, finally retinal blood vessels are extracted with entropy threshold method. Compared to the result segmented by manual of experts, our approach can get more accuracy and lower misclassify ratio

Keywords

blood vessels; diseases; eye; feature extraction; image registration; image segmentation; mathematical morphology; medical image processing; background removal; blood vessel extraction; brightness; diseases; entropy threshold; gray mathematical morphology; image mosaic; image registration; retinal image vessel segmentation; Biomedical imaging; Blood vessels; Diseases; Entropy; Image reconstruction; Image segmentation; Matched filters; Morphology; Pathology; Retina; Vessel segmentation; mathematical morphology; retinal image;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on

Conference_Location

Dalian

Print_ISBN

1-4244-0332-4

Type

conf

DOI

10.1109/WCICA.2006.1713880

Filename

1713880