Exudates segmentation using inverse surface adaptive thresholding

Haniza Yazid, Hamzah Arof, Hazlita Mohd Isa

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

This paper presents a new approach to detect exudates and optic disc from color fundus images based on inverse surface thresholding. The strategy involves the applications of fuzzy c-means clustering, edge detection, otsu thresholding and inverse surface thresholding. The main advantage of the proposed approach is that it does not depend on manually selected parameters that are normally chosen to suit the tested databases. When applied to two sets of databases the proposed method outperforms methods based on watershed segmentation and morphological reconstruction. The proposed method obtained 98.2 and 90.4 in terms of sensitivity for Standard Diabetic Retinopathy Database - Calibration Level 1 (DIARETDB1) and a local dataset provided by National University Hospital of Malaysia (NUHM), respectively.

Original languageEnglish
Pages (from-to)1599-1608
Number of pages10
JournalMeasurement: Journal of the International Measurement Confederation
Volume45
Issue number6
DOIs
Publication statusPublished - Jul 2012

Fingerprint

Adaptive Thresholding
Thresholding
Segmentation
Malaysia
edge detection
Fuzzy C-means Clustering
Edge Detection
Edge detection
optics
Color Image
Watersheds
color
Optics
sensitivity
Calibration
Color

Keywords

  • Biomedical applications
  • Diabetic retinopathy
  • Image segmentation
  • Inverse surface thresholding
  • Medical images
  • Thresholding

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Applied Mathematics

Cite this

Exudates segmentation using inverse surface adaptive thresholding. / Yazid, Haniza; Arof, Hamzah; Mohd Isa, Hazlita.

In: Measurement: Journal of the International Measurement Confederation, Vol. 45, No. 6, 07.2012, p. 1599-1608.

Research output: Contribution to journalArticle

Yazid, Haniza ; Arof, Hamzah ; Mohd Isa, Hazlita. / Exudates segmentation using inverse surface adaptive thresholding. In: Measurement: Journal of the International Measurement Confederation. 2012 ; Vol. 45, No. 6. pp. 1599-1608.
@article{04b94d15afc749ac9edc673328602475,
title = "Exudates segmentation using inverse surface adaptive thresholding",
abstract = "This paper presents a new approach to detect exudates and optic disc from color fundus images based on inverse surface thresholding. The strategy involves the applications of fuzzy c-means clustering, edge detection, otsu thresholding and inverse surface thresholding. The main advantage of the proposed approach is that it does not depend on manually selected parameters that are normally chosen to suit the tested databases. When applied to two sets of databases the proposed method outperforms methods based on watershed segmentation and morphological reconstruction. The proposed method obtained 98.2 and 90.4 in terms of sensitivity for Standard Diabetic Retinopathy Database - Calibration Level 1 (DIARETDB1) and a local dataset provided by National University Hospital of Malaysia (NUHM), respectively.",
keywords = "Biomedical applications, Diabetic retinopathy, Image segmentation, Inverse surface thresholding, Medical images, Thresholding",
author = "Haniza Yazid and Hamzah Arof and {Mohd Isa}, Hazlita",
year = "2012",
month = "7",
doi = "10.1016/j.measurement.2012.02.016",
language = "English",
volume = "45",
pages = "1599--1608",
journal = "Measurement",
issn = "1536-6367",
publisher = "Elsevier",
number = "6",

}

TY - JOUR

T1 - Exudates segmentation using inverse surface adaptive thresholding

AU - Yazid, Haniza

AU - Arof, Hamzah

AU - Mohd Isa, Hazlita

PY - 2012/7

Y1 - 2012/7

N2 - This paper presents a new approach to detect exudates and optic disc from color fundus images based on inverse surface thresholding. The strategy involves the applications of fuzzy c-means clustering, edge detection, otsu thresholding and inverse surface thresholding. The main advantage of the proposed approach is that it does not depend on manually selected parameters that are normally chosen to suit the tested databases. When applied to two sets of databases the proposed method outperforms methods based on watershed segmentation and morphological reconstruction. The proposed method obtained 98.2 and 90.4 in terms of sensitivity for Standard Diabetic Retinopathy Database - Calibration Level 1 (DIARETDB1) and a local dataset provided by National University Hospital of Malaysia (NUHM), respectively.

AB - This paper presents a new approach to detect exudates and optic disc from color fundus images based on inverse surface thresholding. The strategy involves the applications of fuzzy c-means clustering, edge detection, otsu thresholding and inverse surface thresholding. The main advantage of the proposed approach is that it does not depend on manually selected parameters that are normally chosen to suit the tested databases. When applied to two sets of databases the proposed method outperforms methods based on watershed segmentation and morphological reconstruction. The proposed method obtained 98.2 and 90.4 in terms of sensitivity for Standard Diabetic Retinopathy Database - Calibration Level 1 (DIARETDB1) and a local dataset provided by National University Hospital of Malaysia (NUHM), respectively.

KW - Biomedical applications

KW - Diabetic retinopathy

KW - Image segmentation

KW - Inverse surface thresholding

KW - Medical images

KW - Thresholding

UR - http://www.scopus.com/inward/record.url?scp=84860428335&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860428335&partnerID=8YFLogxK

U2 - 10.1016/j.measurement.2012.02.016

DO - 10.1016/j.measurement.2012.02.016

M3 - Article

AN - SCOPUS:84860428335

VL - 45

SP - 1599

EP - 1608

JO - Measurement

JF - Measurement

SN - 1536-6367

IS - 6

ER -