A novel and efficient hybrid segmentation approach for retinal vasculature network

Sidra Aleem, Ping Li, Bin Sheng, Jie Ren

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

1 Citation (Scopus)

Abstract

Segmentation of retinal vasculature network is vital for the timely identification of various eye diseases. In this work, an automated hybrid segmentation technique for retinal vasculature network is proposed. Unlike the other techniques, the proposed method uses separate segmentation techniques for thick and thin retinal vessels. As a result, both vessels are retained in the final segmented result thereby improving the final segmentation accuracy. Enhancement technique used for retinal images can significantly affect the final segmentation result. Rather than using conventional enhancement techniques, for the improved enhancement, Frangi vesselness filter is used. The proposed technique is evaluated on STARE, DRIVE, and a database developed by the authors from the images collected from Shanghai Sixth Peoples Hospital.

Original languageEnglish
Title of host publicationProceedings of the 2018 IEEE International Conference on Progress in Informatics and Computing, PIC 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages191-195
Number of pages5
ISBN (Electronic)9781538676707
DOIs
Publication statusPublished - Dec 2018
Externally publishedYes
Event6th IEEE International Conference on Progress in Informatics and Computing, PIC 2018 - Suzhou, China
Duration: 14 Dec 201816 Dec 2018

Publication series

NameProceedings of the 2018 IEEE International Conference on Progress in Informatics and Computing, PIC 2018

Conference

Conference6th IEEE International Conference on Progress in Informatics and Computing, PIC 2018
Country/TerritoryChina
CitySuzhou
Period14/12/1816/12/18

Keywords

  • Hybrid segmentation
  • retinal image analysis
  • retinal vasculature network

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'A novel and efficient hybrid segmentation approach for retinal vasculature network'. Together they form a unique fingerprint.

Cite this