Weight determination for supervised binarization algorithm based on QR decomposition

Kasmin Fauziah, Azizi Abdullah, Prabuwono Anton Satria

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Supervised binarization is a method that learn pre-classified data in order to classify a particular pixel whether it is belong to a foreground or a background. The performance of supervised approach is usually better than that of unsupervised ones since it is designed to use classification criteria determined by ground truth data. By using this approach, orientations of local neighbourhood grey level information that are based on eight orientations have been developed to characterize a particular pixel. These orientations are combined together since it may reduce the risk of making a particular poor selection of these orientations. In order to ensemble all orientations, heuristic method have been used to determine weights for each orientation. However, determination of weights using heuristic method is not efficient and not enough as it provides incomplete information. Furthermore, these orientations might be influenced by other different factors. This will lead to wrongly assigning weights to a particular orientation. Hence, determination of weights to combine eight orientations to characterize a particular pixel by using QR decomposition method is proposed. By using QR decomposition method, computational complexity is low and weights obtained for each orientation are optimal. In order to test the proposed approach, 21 document images from DIBCO2009 and DIBCO2011 databases and 55 retinal images from DRIVE and STARE databases have been used. The results of the proposed method clearly show significant improvement where higher average accuracy is obtained compared to by using heuristic method.

Original languageEnglish
Pages (from-to)97-106
Number of pages10
JournalJurnal Teknologi
Volume79
Issue number2
DOIs
Publication statusPublished - 1 Feb 2017

Fingerprint

Heuristic methods
Pixels
Decomposition
Computational complexity

Keywords

  • Binarization
  • Ensemble
  • Local neighbourhood
  • QR decomposition method
  • Weights

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Weight determination for supervised binarization algorithm based on QR decomposition. / Fauziah, Kasmin; Abdullah, Azizi; Anton Satria, Prabuwono.

In: Jurnal Teknologi, Vol. 79, No. 2, 01.02.2017, p. 97-106.

Research output: Contribution to journalArticle

Fauziah, Kasmin ; Abdullah, Azizi ; Anton Satria, Prabuwono. / Weight determination for supervised binarization algorithm based on QR decomposition. In: Jurnal Teknologi. 2017 ; Vol. 79, No. 2. pp. 97-106.
@article{66dcb1301ed34a22adc084ba8c7d09ff,
title = "Weight determination for supervised binarization algorithm based on QR decomposition",
abstract = "Supervised binarization is a method that learn pre-classified data in order to classify a particular pixel whether it is belong to a foreground or a background. The performance of supervised approach is usually better than that of unsupervised ones since it is designed to use classification criteria determined by ground truth data. By using this approach, orientations of local neighbourhood grey level information that are based on eight orientations have been developed to characterize a particular pixel. These orientations are combined together since it may reduce the risk of making a particular poor selection of these orientations. In order to ensemble all orientations, heuristic method have been used to determine weights for each orientation. However, determination of weights using heuristic method is not efficient and not enough as it provides incomplete information. Furthermore, these orientations might be influenced by other different factors. This will lead to wrongly assigning weights to a particular orientation. Hence, determination of weights to combine eight orientations to characterize a particular pixel by using QR decomposition method is proposed. By using QR decomposition method, computational complexity is low and weights obtained for each orientation are optimal. In order to test the proposed approach, 21 document images from DIBCO2009 and DIBCO2011 databases and 55 retinal images from DRIVE and STARE databases have been used. The results of the proposed method clearly show significant improvement where higher average accuracy is obtained compared to by using heuristic method.",
keywords = "Binarization, Ensemble, Local neighbourhood, QR decomposition method, Weights",
author = "Kasmin Fauziah and Azizi Abdullah and {Anton Satria}, Prabuwono",
year = "2017",
month = "2",
day = "1",
doi = "10.11113/jt.v79.7185",
language = "English",
volume = "79",
pages = "97--106",
journal = "Jurnal Teknologi",
issn = "0127-9696",
publisher = "Penerbit Universiti Teknologi Malaysia",
number = "2",

}

TY - JOUR

T1 - Weight determination for supervised binarization algorithm based on QR decomposition

AU - Fauziah, Kasmin

AU - Abdullah, Azizi

AU - Anton Satria, Prabuwono

PY - 2017/2/1

Y1 - 2017/2/1

N2 - Supervised binarization is a method that learn pre-classified data in order to classify a particular pixel whether it is belong to a foreground or a background. The performance of supervised approach is usually better than that of unsupervised ones since it is designed to use classification criteria determined by ground truth data. By using this approach, orientations of local neighbourhood grey level information that are based on eight orientations have been developed to characterize a particular pixel. These orientations are combined together since it may reduce the risk of making a particular poor selection of these orientations. In order to ensemble all orientations, heuristic method have been used to determine weights for each orientation. However, determination of weights using heuristic method is not efficient and not enough as it provides incomplete information. Furthermore, these orientations might be influenced by other different factors. This will lead to wrongly assigning weights to a particular orientation. Hence, determination of weights to combine eight orientations to characterize a particular pixel by using QR decomposition method is proposed. By using QR decomposition method, computational complexity is low and weights obtained for each orientation are optimal. In order to test the proposed approach, 21 document images from DIBCO2009 and DIBCO2011 databases and 55 retinal images from DRIVE and STARE databases have been used. The results of the proposed method clearly show significant improvement where higher average accuracy is obtained compared to by using heuristic method.

AB - Supervised binarization is a method that learn pre-classified data in order to classify a particular pixel whether it is belong to a foreground or a background. The performance of supervised approach is usually better than that of unsupervised ones since it is designed to use classification criteria determined by ground truth data. By using this approach, orientations of local neighbourhood grey level information that are based on eight orientations have been developed to characterize a particular pixel. These orientations are combined together since it may reduce the risk of making a particular poor selection of these orientations. In order to ensemble all orientations, heuristic method have been used to determine weights for each orientation. However, determination of weights using heuristic method is not efficient and not enough as it provides incomplete information. Furthermore, these orientations might be influenced by other different factors. This will lead to wrongly assigning weights to a particular orientation. Hence, determination of weights to combine eight orientations to characterize a particular pixel by using QR decomposition method is proposed. By using QR decomposition method, computational complexity is low and weights obtained for each orientation are optimal. In order to test the proposed approach, 21 document images from DIBCO2009 and DIBCO2011 databases and 55 retinal images from DRIVE and STARE databases have been used. The results of the proposed method clearly show significant improvement where higher average accuracy is obtained compared to by using heuristic method.

KW - Binarization

KW - Ensemble

KW - Local neighbourhood

KW - QR decomposition method

KW - Weights

UR - http://www.scopus.com/inward/record.url?scp=85011706167&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011706167&partnerID=8YFLogxK

U2 - 10.11113/jt.v79.7185

DO - 10.11113/jt.v79.7185

M3 - Article

VL - 79

SP - 97

EP - 106

JO - Jurnal Teknologi

JF - Jurnal Teknologi

SN - 0127-9696

IS - 2

ER -