Automatic Blood Vessel Segmentation in Retinal Image Based on Mathematical Morphology
Vandana Y. Koli1, Ashwini G. Andurkar2, Harsha S. Jain3
1Miss. Vandana Koli, Department of Electronics & Tele Communication, Government College of Engineering, Jalgaon, India.
2Mrs. Ashwini Andurkar, Department of Electronics & Tele Communication, Government College of Engineering, Jalgaon, India.
3Miss. Harsha Jain, Department of Electronics & Tele Communication, Government College of Engineering, Jalgaon, India.
Manuscript received on November 06, 2014. | Revised Manuscript Received on November 20, 2014. | Manuscript published on November 20, 2014. | PP: 33-37 | Volume-2, Issue-12, November 2014. | Retrieval Number: L05471121214/2014©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Retinal blood vessels detection or segmentation is important according to ophthalmologist. To diagnose the retinal disease or to avoid the vision loss, regular checkup of retinal blood vessels is necessary. This regular checkup provides the information about the changes of blood vessels. This changes are like swelling, narrowing of blood vessels etc. The automatic segmentation of blood vessels helps in the diagnosis of retinal diseases. In this work two approaches are used for vessel segmentation. First one is segmentation using morphology with Thresholding and second is segmentation using morphology with Fuzzy-C-Means clustering. Both approaches are unsupervised methods. The segmentation result of these methods is approximately same but there is one difference. The first one technique provides better result for major vessel while second provides good result for minor vessels. This system designed to resolve the problem of ophthalmologist by developing two algorithms.
Keywords: Retinal Blood Vessels, Fuzzy-C-Means, Mathematical Morphology, Thresholding