Publication detail

Classification-Based Blood Vessel Segmentation in Retinal Images

ODSTRČILÍK, J. KOLÁŘ, R. HARABIŠ, V. TORNOW, R.

Original Title

Classification-Based Blood Vessel Segmentation in Retinal Images

English Title

Classification-Based Blood Vessel Segmentation in Retinal Images

Type

conference paper

Language

en

Original Abstract

Automatic and precise segmentation of retinal blood vessels can help in computer aided as-sessment of various retinal diseases, especially diseases related to cardiovascular system or glaucoma. Hence, an accurate detection of retinal vascular structures is one of the most addressed topic in the field of retinal im-age processing today. Most of the methods are designed directly for utilization with a special dataset only or have problems to segment blurry and noisy images. In this study, we showed that combination of three stand-ard approaches – matched filtering, Hessian-based approach, and morphological processing, together with support vector machine classification technique, can be satisfactorily used as a universal segmentation ap-proach in retinal images acquired using entirely different devices.

English abstract

Automatic and precise segmentation of retinal blood vessels can help in computer aided as-sessment of various retinal diseases, especially diseases related to cardiovascular system or glaucoma. Hence, an accurate detection of retinal vascular structures is one of the most addressed topic in the field of retinal im-age processing today. Most of the methods are designed directly for utilization with a special dataset only or have problems to segment blurry and noisy images. In this study, we showed that combination of three stand-ard approaches – matched filtering, Hessian-based approach, and morphological processing, together with support vector machine classification technique, can be satisfactorily used as a universal segmentation ap-proach in retinal images acquired using entirely different devices.

Keywords

segmentation, retinal vessels, blood vessels, classification, retinal images

RIV year

2015

Released

19.10.2015

Publisher

CRC Press Taylor and Francis Group

Location

London, UK

ISBN

978-1-138-02926-2

Book

Computational Vision and Medical Image Processing V

Pages from

95

Pages to

100

Pages count

6

BibTex


@inproceedings{BUT117935,
  author="Jan {Odstrčilík} and Radim {Kolář} and Vratislav {Harabiš} and Ralf-Peter {Tornow}",
  title="Classification-Based Blood Vessel Segmentation in Retinal Images",
  annote="Automatic and precise segmentation of retinal blood vessels can help in computer aided as-sessment of various retinal diseases, especially diseases related to cardiovascular system or glaucoma. Hence, an accurate detection of retinal vascular structures is one of the most addressed topic in the field of retinal im-age processing today. Most of the methods are designed directly for utilization with a special dataset only or have problems to segment blurry and noisy images. In this study, we showed that combination of three stand-ard approaches – matched filtering, Hessian-based approach, and morphological processing, together with support vector machine classification technique, can be satisfactorily used as a universal segmentation ap-proach in retinal images acquired using entirely different devices.",
  address="CRC Press Taylor and Francis Group",
  booktitle="Computational Vision and Medical Image Processing V",
  chapter="117935",
  howpublished="print",
  institution="CRC Press Taylor and Francis Group",
  year="2015",
  month="october",
  pages="95--100",
  publisher="CRC Press Taylor and Francis Group",
  type="conference paper"
}