We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Research on Segmentation Algorithms of Retinal Vessel Images.
- Authors
Hao-Ming Song; Yu Liu; Jie-Sheng Wang; Biao Zhou
- Abstract
Retinal diabetes, hypertension, cardiovascular and cerebrovascular diseases are mostly diagnosed automatically in medicine, and retinal vascular segmentation results have a great impact on them. Because of the tiny blood vessels, pathological changes, central reflection and uneven background brightness in retinal images, the accuracy of blood vessel segmentation is low. Because the illumination of blood vessel pixels and background pixels in fundus image is uneven and the contrast is low. In this paper, data standardization, mathematical morphology, adaptive histogram equalization, gray conversion and gamma correction are used to carry out preparatory processing on images of retinal blood vessels, which can not only detect the characteristics of small blood vessels more accurately, but also improve the performance of network segmentation. Then, the retinal vessel images are segmented based on six segmentation algorithms: maximum variance between classes, region growing, Sobel operator, iterative threshold method, quad-tree image segmentation and mathematical morphology algorithm. Simulation experiments are carried out to verify the effectiveness of the segmentation algorithms for typical retinal vessel images in Drive library.
- Subjects
RETINAL blood vessels; RETINAL imaging; BLOOD vessels; MATHEMATICAL morphology; IMAGE segmentation; CEREBROVASCULAR disease
- Publication
IAENG International Journal of Computer Science, 2022, Vol 49, Issue 2, p286
- ISSN
1819-656X
- Publication type
Article