Adaptive binarization method for degraded document images based on surface contrast variation

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Document binarization is an important technique in document image analysis and recognition. Generally, binarization methods are ineffective for degraded images. Several binarization methods have been proposed; however, none of them are effective for historical and degraded document images. In this paper, a new binarization method is proposed for degraded document images. The proposed method based on the variance between pixel contrast, it consists of four stages: pre-processing, geometrical feature extraction, feature selection, and post-processing. The proposed method was evaluated based on several visual and statistical experiments. The experiments were conducted using five International Document Image Binarization Contest benchmark datasets specialized for binarization testing. The results compared with five adaptive binarization methods: Niblack, Sauvola thresholding, Sauvola compound algorithm, NICK, and Bataineh. The results show that the proposed method performs better than other methods in all binarization cases.

Original languageEnglish
JournalPattern Analysis and Applications
DOIs
Publication statusAccepted/In press - 30 Oct 2015

Fingerprint

Feature extraction
Image recognition
Processing
Image analysis
Pixels
Experiments
Testing

Keywords

  • Degraded images
  • Document image binarization
  • Historical images
  • Laplacian filter
  • OCR
  • OFF center-surround cells

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition

Cite this

@article{9b1cc429d74c4f2c8889d77076d5980c,
title = "Adaptive binarization method for degraded document images based on surface contrast variation",
abstract = "Document binarization is an important technique in document image analysis and recognition. Generally, binarization methods are ineffective for degraded images. Several binarization methods have been proposed; however, none of them are effective for historical and degraded document images. In this paper, a new binarization method is proposed for degraded document images. The proposed method based on the variance between pixel contrast, it consists of four stages: pre-processing, geometrical feature extraction, feature selection, and post-processing. The proposed method was evaluated based on several visual and statistical experiments. The experiments were conducted using five International Document Image Binarization Contest benchmark datasets specialized for binarization testing. The results compared with five adaptive binarization methods: Niblack, Sauvola thresholding, Sauvola compound algorithm, NICK, and Bataineh. The results show that the proposed method performs better than other methods in all binarization cases.",
keywords = "Degraded images, Document image binarization, Historical images, Laplacian filter, OCR, OFF center-surround cells",
author = "Bilal Bataineh and {Sheikh Abdullah}, {Siti Norul Huda} and Khairuddin Omar",
year = "2015",
month = "10",
day = "30",
doi = "10.1007/s10044-015-0520-0",
language = "English",
journal = "Pattern Analysis and Applications",
issn = "1433-7541",
publisher = "Springer London",

}

TY - JOUR

T1 - Adaptive binarization method for degraded document images based on surface contrast variation

AU - Bataineh, Bilal

AU - Sheikh Abdullah, Siti Norul Huda

AU - Omar, Khairuddin

PY - 2015/10/30

Y1 - 2015/10/30

N2 - Document binarization is an important technique in document image analysis and recognition. Generally, binarization methods are ineffective for degraded images. Several binarization methods have been proposed; however, none of them are effective for historical and degraded document images. In this paper, a new binarization method is proposed for degraded document images. The proposed method based on the variance between pixel contrast, it consists of four stages: pre-processing, geometrical feature extraction, feature selection, and post-processing. The proposed method was evaluated based on several visual and statistical experiments. The experiments were conducted using five International Document Image Binarization Contest benchmark datasets specialized for binarization testing. The results compared with five adaptive binarization methods: Niblack, Sauvola thresholding, Sauvola compound algorithm, NICK, and Bataineh. The results show that the proposed method performs better than other methods in all binarization cases.

AB - Document binarization is an important technique in document image analysis and recognition. Generally, binarization methods are ineffective for degraded images. Several binarization methods have been proposed; however, none of them are effective for historical and degraded document images. In this paper, a new binarization method is proposed for degraded document images. The proposed method based on the variance between pixel contrast, it consists of four stages: pre-processing, geometrical feature extraction, feature selection, and post-processing. The proposed method was evaluated based on several visual and statistical experiments. The experiments were conducted using five International Document Image Binarization Contest benchmark datasets specialized for binarization testing. The results compared with five adaptive binarization methods: Niblack, Sauvola thresholding, Sauvola compound algorithm, NICK, and Bataineh. The results show that the proposed method performs better than other methods in all binarization cases.

KW - Degraded images

KW - Document image binarization

KW - Historical images

KW - Laplacian filter

KW - OCR

KW - OFF center-surround cells

UR - http://www.scopus.com/inward/record.url?scp=84945540327&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945540327&partnerID=8YFLogxK

U2 - 10.1007/s10044-015-0520-0

DO - 10.1007/s10044-015-0520-0

M3 - Article

AN - SCOPUS:84945540327

JO - Pattern Analysis and Applications

JF - Pattern Analysis and Applications

SN - 1433-7541

ER -