Ensembles of normalization techniques to improve the accuracy of otsu method

Fauziah Kasmin, Azizi Abdullah, Anton Satria Prabuwono

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Otsu method is a global thresholding method that uses between class variance as a discriminant criterion in order to maximize the separation between background and foreground of an image. However, there are problems of biasness in Otsu method. These problems are caused by the differences in class variances. The threshold value obtained by Otsu method will be bias towards the larger class variance component. Hence, in this paper, a new variant of Otsu method by using normalization techniques and their ensembles is proposed. By using normalization techniques, grey level values will be transformed into a smaller range in feature space and this will affect Otsu method as this method depends on grey level values. The domination of certain grey level values also will be eliminated. Rank filtering has been applied to eliminate noises and ensemble approaches of normalization techniques are utilized to increase the performance of the proposed method. Ensemble approaches namely Maximum Variance, Majority Voting, Product Rule, Addition Rule and Average Rule have been applied on the binary images obtained. From the experiment on 20 retinal images randomly selected in 50 runs from DRIVE and STARE databases, Maximum Variance shows the most significant result that is 95.39% accuracy. While from the experiment on 15 document images randomly selected in 50 runs from DIBCO2009 and DIBCO2011 databases, Average Rule shows the most significant result that is 97.17% accuracy. This indicate the use of ensembles of normalization techniques can give promising result to improve Otsu method.

Original languageEnglish
Pages (from-to)1565-1578
Number of pages14
JournalApplied Mathematical Sciences
Volume9
Issue number29-32
DOIs
Publication statusPublished - 2015

Fingerprint

Normalization
Ensemble
Binary images
Experiments
Product rule
Majority Voting
Variance Components
Binary Image
Thresholding
Domination
Feature Space
Threshold Value
Discriminant
Experiment
Eliminate
Filtering
Maximise
Range of data
Class

Keywords

  • Ensemble approaches
  • Normalization techniques
  • Otsu method
  • Segmentation
  • Thresholding

ASJC Scopus subject areas

  • Applied Mathematics

Cite this

Ensembles of normalization techniques to improve the accuracy of otsu method. / Kasmin, Fauziah; Abdullah, Azizi; Prabuwono, Anton Satria.

In: Applied Mathematical Sciences, Vol. 9, No. 29-32, 2015, p. 1565-1578.

Research output: Contribution to journalArticle

Kasmin, Fauziah ; Abdullah, Azizi ; Prabuwono, Anton Satria. / Ensembles of normalization techniques to improve the accuracy of otsu method. In: Applied Mathematical Sciences. 2015 ; Vol. 9, No. 29-32. pp. 1565-1578.
@article{bd4e5f9a949444ea900bb0ef92c94a33,
title = "Ensembles of normalization techniques to improve the accuracy of otsu method",
abstract = "Otsu method is a global thresholding method that uses between class variance as a discriminant criterion in order to maximize the separation between background and foreground of an image. However, there are problems of biasness in Otsu method. These problems are caused by the differences in class variances. The threshold value obtained by Otsu method will be bias towards the larger class variance component. Hence, in this paper, a new variant of Otsu method by using normalization techniques and their ensembles is proposed. By using normalization techniques, grey level values will be transformed into a smaller range in feature space and this will affect Otsu method as this method depends on grey level values. The domination of certain grey level values also will be eliminated. Rank filtering has been applied to eliminate noises and ensemble approaches of normalization techniques are utilized to increase the performance of the proposed method. Ensemble approaches namely Maximum Variance, Majority Voting, Product Rule, Addition Rule and Average Rule have been applied on the binary images obtained. From the experiment on 20 retinal images randomly selected in 50 runs from DRIVE and STARE databases, Maximum Variance shows the most significant result that is 95.39{\%} accuracy. While from the experiment on 15 document images randomly selected in 50 runs from DIBCO2009 and DIBCO2011 databases, Average Rule shows the most significant result that is 97.17{\%} accuracy. This indicate the use of ensembles of normalization techniques can give promising result to improve Otsu method.",
keywords = "Ensemble approaches, Normalization techniques, Otsu method, Segmentation, Thresholding",
author = "Fauziah Kasmin and Azizi Abdullah and Prabuwono, {Anton Satria}",
year = "2015",
doi = "10.12988/ams.2015.517",
language = "English",
volume = "9",
pages = "1565--1578",
journal = "Applied Mathematical Sciences",
issn = "1312-885X",
publisher = "Hikari Ltd.",
number = "29-32",

}

TY - JOUR

T1 - Ensembles of normalization techniques to improve the accuracy of otsu method

AU - Kasmin, Fauziah

AU - Abdullah, Azizi

AU - Prabuwono, Anton Satria

PY - 2015

Y1 - 2015

N2 - Otsu method is a global thresholding method that uses between class variance as a discriminant criterion in order to maximize the separation between background and foreground of an image. However, there are problems of biasness in Otsu method. These problems are caused by the differences in class variances. The threshold value obtained by Otsu method will be bias towards the larger class variance component. Hence, in this paper, a new variant of Otsu method by using normalization techniques and their ensembles is proposed. By using normalization techniques, grey level values will be transformed into a smaller range in feature space and this will affect Otsu method as this method depends on grey level values. The domination of certain grey level values also will be eliminated. Rank filtering has been applied to eliminate noises and ensemble approaches of normalization techniques are utilized to increase the performance of the proposed method. Ensemble approaches namely Maximum Variance, Majority Voting, Product Rule, Addition Rule and Average Rule have been applied on the binary images obtained. From the experiment on 20 retinal images randomly selected in 50 runs from DRIVE and STARE databases, Maximum Variance shows the most significant result that is 95.39% accuracy. While from the experiment on 15 document images randomly selected in 50 runs from DIBCO2009 and DIBCO2011 databases, Average Rule shows the most significant result that is 97.17% accuracy. This indicate the use of ensembles of normalization techniques can give promising result to improve Otsu method.

AB - Otsu method is a global thresholding method that uses between class variance as a discriminant criterion in order to maximize the separation between background and foreground of an image. However, there are problems of biasness in Otsu method. These problems are caused by the differences in class variances. The threshold value obtained by Otsu method will be bias towards the larger class variance component. Hence, in this paper, a new variant of Otsu method by using normalization techniques and their ensembles is proposed. By using normalization techniques, grey level values will be transformed into a smaller range in feature space and this will affect Otsu method as this method depends on grey level values. The domination of certain grey level values also will be eliminated. Rank filtering has been applied to eliminate noises and ensemble approaches of normalization techniques are utilized to increase the performance of the proposed method. Ensemble approaches namely Maximum Variance, Majority Voting, Product Rule, Addition Rule and Average Rule have been applied on the binary images obtained. From the experiment on 20 retinal images randomly selected in 50 runs from DRIVE and STARE databases, Maximum Variance shows the most significant result that is 95.39% accuracy. While from the experiment on 15 document images randomly selected in 50 runs from DIBCO2009 and DIBCO2011 databases, Average Rule shows the most significant result that is 97.17% accuracy. This indicate the use of ensembles of normalization techniques can give promising result to improve Otsu method.

KW - Ensemble approaches

KW - Normalization techniques

KW - Otsu method

KW - Segmentation

KW - Thresholding

UR - http://www.scopus.com/inward/record.url?scp=84929892944&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929892944&partnerID=8YFLogxK

U2 - 10.12988/ams.2015.517

DO - 10.12988/ams.2015.517

M3 - Article

AN - SCOPUS:84929892944

VL - 9

SP - 1565

EP - 1578

JO - Applied Mathematical Sciences

JF - Applied Mathematical Sciences

SN - 1312-885X

IS - 29-32

ER -