Extraction of blood vessels in retinal images using resampling high-order background estimation

Sukritta Paripurana, Werapon Chiracharit, Kosin Chamnongthai, Hideo Saito

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In retinal blood vessel extraction through background removal, the vessels in a fundus image which appear in a higher illumination variance area are often missing after the background is removed. This is because the intensity values of the vessel and the background are nearly the same. Thus, the estimated background should be robust to changes of the illumination intensity. This paper proposes retinal blood vessel extraction using background estimation. The estimated background is calculated by using a weight surface fitting method with a high degree polynomial. Bright pixels are defined as unwanted data and are set as zero in a weight matrix. To fit a retinal surface with a higher degree polynomial, fundus images are reduced in size by different scaling parameters in order to reduce the processing time and complexity in calculation. The estimated background is then removed from the original image. The candidate vessel pixels are extracted from the image by using the local threshold values. To identify the true vessel region, the candidate vessel pixels are dilated from the candidate. After that, the active contour without edge method is applied. The experimental results show that the efficiency of the proposed method is higher than the conventional low-pass filter and the conventional surface fitting method. Moreover, rescaling an image down using the scaling parameter at 0.25 before background estimation provides as good a result as a non-rescaled image does. The correlation value between the non-rescaled image and the rescaled image is 0.99. The results of the proposed method in the sensitivity, the specificity, the accuracy, the area under the receiver operating characteristic (ROC) curve (AUC) and the processing time per image are 0.7994, 0.9717, 0.9543, 0.9676 and 1.8320 seconds for the DRIVE database respectively.

Original languageEnglish
Pages (from-to)692-703
Number of pages12
JournalIEICE Transactions on Information and Systems
VolumeE98D
Issue number3
DOIs
Publication statusPublished - 2015 Mar 1

Fingerprint

Blood vessels
Pixels
Lighting
Polynomials
Low pass filters
Processing

Keywords

  • High order degree polynomial
  • Rescaling image
  • Retinal background estimation
  • Retinal blood vessel extraction

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Extraction of blood vessels in retinal images using resampling high-order background estimation. / Paripurana, Sukritta; Chiracharit, Werapon; Chamnongthai, Kosin; Saito, Hideo.

In: IEICE Transactions on Information and Systems, Vol. E98D, No. 3, 01.03.2015, p. 692-703.

Research output: Contribution to journalArticle

Paripurana, Sukritta ; Chiracharit, Werapon ; Chamnongthai, Kosin ; Saito, Hideo. / Extraction of blood vessels in retinal images using resampling high-order background estimation. In: IEICE Transactions on Information and Systems. 2015 ; Vol. E98D, No. 3. pp. 692-703.
@article{67fd3cd1403e4288852303e84e098bf7,
title = "Extraction of blood vessels in retinal images using resampling high-order background estimation",
abstract = "In retinal blood vessel extraction through background removal, the vessels in a fundus image which appear in a higher illumination variance area are often missing after the background is removed. This is because the intensity values of the vessel and the background are nearly the same. Thus, the estimated background should be robust to changes of the illumination intensity. This paper proposes retinal blood vessel extraction using background estimation. The estimated background is calculated by using a weight surface fitting method with a high degree polynomial. Bright pixels are defined as unwanted data and are set as zero in a weight matrix. To fit a retinal surface with a higher degree polynomial, fundus images are reduced in size by different scaling parameters in order to reduce the processing time and complexity in calculation. The estimated background is then removed from the original image. The candidate vessel pixels are extracted from the image by using the local threshold values. To identify the true vessel region, the candidate vessel pixels are dilated from the candidate. After that, the active contour without edge method is applied. The experimental results show that the efficiency of the proposed method is higher than the conventional low-pass filter and the conventional surface fitting method. Moreover, rescaling an image down using the scaling parameter at 0.25 before background estimation provides as good a result as a non-rescaled image does. The correlation value between the non-rescaled image and the rescaled image is 0.99. The results of the proposed method in the sensitivity, the specificity, the accuracy, the area under the receiver operating characteristic (ROC) curve (AUC) and the processing time per image are 0.7994, 0.9717, 0.9543, 0.9676 and 1.8320 seconds for the DRIVE database respectively.",
keywords = "High order degree polynomial, Rescaling image, Retinal background estimation, Retinal blood vessel extraction",
author = "Sukritta Paripurana and Werapon Chiracharit and Kosin Chamnongthai and Hideo Saito",
year = "2015",
month = "3",
day = "1",
doi = "10.1587/transinf.2014EDP7186",
language = "English",
volume = "E98D",
pages = "692--703",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "3",

}

TY - JOUR

T1 - Extraction of blood vessels in retinal images using resampling high-order background estimation

AU - Paripurana, Sukritta

AU - Chiracharit, Werapon

AU - Chamnongthai, Kosin

AU - Saito, Hideo

PY - 2015/3/1

Y1 - 2015/3/1

N2 - In retinal blood vessel extraction through background removal, the vessels in a fundus image which appear in a higher illumination variance area are often missing after the background is removed. This is because the intensity values of the vessel and the background are nearly the same. Thus, the estimated background should be robust to changes of the illumination intensity. This paper proposes retinal blood vessel extraction using background estimation. The estimated background is calculated by using a weight surface fitting method with a high degree polynomial. Bright pixels are defined as unwanted data and are set as zero in a weight matrix. To fit a retinal surface with a higher degree polynomial, fundus images are reduced in size by different scaling parameters in order to reduce the processing time and complexity in calculation. The estimated background is then removed from the original image. The candidate vessel pixels are extracted from the image by using the local threshold values. To identify the true vessel region, the candidate vessel pixels are dilated from the candidate. After that, the active contour without edge method is applied. The experimental results show that the efficiency of the proposed method is higher than the conventional low-pass filter and the conventional surface fitting method. Moreover, rescaling an image down using the scaling parameter at 0.25 before background estimation provides as good a result as a non-rescaled image does. The correlation value between the non-rescaled image and the rescaled image is 0.99. The results of the proposed method in the sensitivity, the specificity, the accuracy, the area under the receiver operating characteristic (ROC) curve (AUC) and the processing time per image are 0.7994, 0.9717, 0.9543, 0.9676 and 1.8320 seconds for the DRIVE database respectively.

AB - In retinal blood vessel extraction through background removal, the vessels in a fundus image which appear in a higher illumination variance area are often missing after the background is removed. This is because the intensity values of the vessel and the background are nearly the same. Thus, the estimated background should be robust to changes of the illumination intensity. This paper proposes retinal blood vessel extraction using background estimation. The estimated background is calculated by using a weight surface fitting method with a high degree polynomial. Bright pixels are defined as unwanted data and are set as zero in a weight matrix. To fit a retinal surface with a higher degree polynomial, fundus images are reduced in size by different scaling parameters in order to reduce the processing time and complexity in calculation. The estimated background is then removed from the original image. The candidate vessel pixels are extracted from the image by using the local threshold values. To identify the true vessel region, the candidate vessel pixels are dilated from the candidate. After that, the active contour without edge method is applied. The experimental results show that the efficiency of the proposed method is higher than the conventional low-pass filter and the conventional surface fitting method. Moreover, rescaling an image down using the scaling parameter at 0.25 before background estimation provides as good a result as a non-rescaled image does. The correlation value between the non-rescaled image and the rescaled image is 0.99. The results of the proposed method in the sensitivity, the specificity, the accuracy, the area under the receiver operating characteristic (ROC) curve (AUC) and the processing time per image are 0.7994, 0.9717, 0.9543, 0.9676 and 1.8320 seconds for the DRIVE database respectively.

KW - High order degree polynomial

KW - Rescaling image

KW - Retinal background estimation

KW - Retinal blood vessel extraction

UR - http://www.scopus.com/inward/record.url?scp=84923853739&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923853739&partnerID=8YFLogxK

U2 - 10.1587/transinf.2014EDP7186

DO - 10.1587/transinf.2014EDP7186

M3 - Article

VL - E98D

SP - 692

EP - 703

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 3

ER -