Publication detail

Automatic detection of the macula in retinal fundus images using multilevel thresholding

MINÁŘ, J. ŘÍHA, K. KŘUPKA, A. TONG, H.

Original Title

Automatic detection of the macula in retinal fundus images using multilevel thresholding

English Title

Automatic detection of the macula in retinal fundus images using multilevel thresholding

Type

journal article - other

Language

en

Original Abstract

The paper proposes a novel method for detection of macula in medical image of human eye – retinal fundus images that can be used in ophthalmology for detecting various eye's diseases such glaucoma, diabetic retinopathy or macula oedema. The method utilizes an approach of multilevel thresholding of red channel of fundus retinal image. Subsequently, the thresholded layers are preprocessed by application of median blur filter. Algorithm for ellipse detection from OpenCV library is used on all thresholding level. Then proposed technique analyzes detection and valuates position of macula in fundus image. The precision of the method is evaluated on dataset from public fundus image library DRIVE. The results were 90% (36 of 40). False detections ware only in images where macula was not visible.

English abstract

The paper proposes a novel method for detection of macula in medical image of human eye – retinal fundus images that can be used in ophthalmology for detecting various eye's diseases such glaucoma, diabetic retinopathy or macula oedema. The method utilizes an approach of multilevel thresholding of red channel of fundus retinal image. Subsequently, the thresholded layers are preprocessed by application of median blur filter. Algorithm for ellipse detection from OpenCV library is used on all thresholding level. Then proposed technique analyzes detection and valuates position of macula in fundus image. The precision of the method is evaluated on dataset from public fundus image library DRIVE. The results were 90% (36 of 40). False detections ware only in images where macula was not visible.

Keywords

Automatic, Fundus, Macula, Medical Image processing, Multilevel Segmentation, Thresholding

RIV year

2014

Released

25.04.2014

ISBN

1805-5443

Periodical

International Journal of Advances in Telecommunications, Electrotechnics, Signals and Systems

Year of study

2014

Number

1

State

CZ

Pages from

13

Pages to

16

Pages count

4

Documents

BibTex


@article{BUT107256,
  author="Jiří {Minář} and Kamil {Říha} and Aleš {Křupka} and Hejun {Tong}",
  title="Automatic detection of the macula in retinal fundus images using multilevel thresholding",
  annote="The paper proposes a novel method for detection of macula in medical image of human eye – retinal fundus images that can be used in ophthalmology for detecting various eye's diseases such glaucoma, diabetic retinopathy or macula oedema. The method utilizes an approach of multilevel thresholding of red channel of fundus retinal image. Subsequently, the thresholded layers are preprocessed by application of median blur filter. Algorithm for ellipse detection from OpenCV library is used on all thresholding level. Then proposed technique analyzes detection and valuates position of macula in fundus image. The precision of the method is evaluated on dataset from public fundus image library DRIVE. The results were 90% (36 of 40). False detections ware only in images where macula was not visible.",
  chapter="107256",
  number="1",
  volume="2014",
  year="2014",
  month="april",
  pages="13--16",
  type="journal article - other"
}