Retinal Blood Vessel Segmentation Using a Generalized Gamma Probability Distribution Function (PDF) of Matched Filtered
Author:
Affiliation:
1. National Institute of Technology, Hamirpur, India
2. National Institute of Technology, Hamirpur, India
Abstract
Retinal images contain information about the retina's blood vessel structure to predict retinal diseases such as diabetics, obesity, glaucoma, etc. Segmentation of accurate retinal blood vessels is a challenging task in the low background of retinal images. Therefore, we proposed a Generalized Gamma Distribution probability distribution function (pdf) to extract the accurate vascular structure on the retinal images. The proposed approach is divided into processing steps, the Generalized Gamma distribution kernel, and the postprocessing step. In pre-processing, the conversion of a color retinal image into a grayscale image using PCA followed by the CLAHE method and the Toggle Contrast method enhances the grayscale images of the retina. The proposed matched filter of Generalized Gamma distribution generates the MFR images. The postprocessing step extracts the thick vessels and thin retinal blood vessels using the optimal thresholding technique. The results obtained on DIRVE database average accuracy 95.00% and the STARE database 93.85%, respectively.
Publisher
IGI Global
Subject
General Computer Science
Reference31 articles.
1. An improved matched filter for blood vessel detection of digital retinal images
2. High speed detection of retinal blood vessels in fundus image using phase congruency
3. Multiscale toggle contrast operator-based mineral image enhancement
4. Detection of blood vessels in retinal images using two-dimensional matched filters
5. Detection of blood vessels in ophthalmoscope images using MF/ant (matched filter/ant colony) algorithm
Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Suppression effects of different extinguishing agents on vent gases fires from lithium-ion batteries after thermal runaway: A comprehensive experimental and numerical study;Fire Safety Journal;2024-09
2. Systematic Review of Retinal Blood Vessels Segmentation Based on AI-driven Technique;Journal of Imaging Informatics in Medicine;2024-03-04
3. Topic Modeling, Sentiment Analysis and Text Summarization for Analyzing News Headlines and Articles;Communications in Computer and Information Science;2022
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3