Markov random field regularisation models for adaptive binarisation of nonuniform images

Dinggang Shen, H. H S Ip

Research output: Chapter in Book/Report/Conference proceedingChapter

4 Citations (Scopus)

Abstract

Two related MRF models, an edge-preserving smoothing model followed by a modified standard regularisation, are presented for the adaptive binarisation of nonuniform images in the presence of noise. In particular, a computational model is developed for a modified standard regularisation method which calculates the adaptive threshold surface for noisy images. Since the modified standard regularisation depends only on the image data, and not its edge segments, it gives much better performance and can be applied to more classes of image than those methods that solely rely on edge segments. Experimental results demonstrate that the proposed method has the best performance over three other commonly used adaptive segmentation methods and is faster than previous interpolation-based thresholding techniques.

Original languageEnglish
Title of host publicationIEE Proceedings: Vision, Image and Signal Processing
Pages322-332
Number of pages11
Volume145
Edition5
Publication statusPublished - 1998
Externally publishedYes

Fingerprint

Interpolation

Keywords

  • Adaptive image binarisation
  • Edge-preserving smoothing
  • Markov random field models
  • Standard regularisation
  • Threshold surface

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing

Cite this

Shen, D., & Ip, H. H. S. (1998). Markov random field regularisation models for adaptive binarisation of nonuniform images. In IEE Proceedings: Vision, Image and Signal Processing (5 ed., Vol. 145, pp. 322-332)

Markov random field regularisation models for adaptive binarisation of nonuniform images. / Shen, Dinggang; Ip, H. H S.

IEE Proceedings: Vision, Image and Signal Processing. Vol. 145 5. ed. 1998. p. 322-332.

Research output: Chapter in Book/Report/Conference proceedingChapter

Shen, D & Ip, HHS 1998, Markov random field regularisation models for adaptive binarisation of nonuniform images. in IEE Proceedings: Vision, Image and Signal Processing. 5 edn, vol. 145, pp. 322-332.
Shen D, Ip HHS. Markov random field regularisation models for adaptive binarisation of nonuniform images. In IEE Proceedings: Vision, Image and Signal Processing. 5 ed. Vol. 145. 1998. p. 322-332
Shen, Dinggang ; Ip, H. H S. / Markov random field regularisation models for adaptive binarisation of nonuniform images. IEE Proceedings: Vision, Image and Signal Processing. Vol. 145 5. ed. 1998. pp. 322-332
@inbook{8d2cc6416f37450393b0b713b2cf2e1f,
title = "Markov random field regularisation models for adaptive binarisation of nonuniform images",
abstract = "Two related MRF models, an edge-preserving smoothing model followed by a modified standard regularisation, are presented for the adaptive binarisation of nonuniform images in the presence of noise. In particular, a computational model is developed for a modified standard regularisation method which calculates the adaptive threshold surface for noisy images. Since the modified standard regularisation depends only on the image data, and not its edge segments, it gives much better performance and can be applied to more classes of image than those methods that solely rely on edge segments. Experimental results demonstrate that the proposed method has the best performance over three other commonly used adaptive segmentation methods and is faster than previous interpolation-based thresholding techniques.",
keywords = "Adaptive image binarisation, Edge-preserving smoothing, Markov random field models, Standard regularisation, Threshold surface",
author = "Dinggang Shen and Ip, {H. H S}",
year = "1998",
language = "English",
volume = "145",
pages = "322--332",
booktitle = "IEE Proceedings: Vision, Image and Signal Processing",
edition = "5",

}

TY - CHAP

T1 - Markov random field regularisation models for adaptive binarisation of nonuniform images

AU - Shen, Dinggang

AU - Ip, H. H S

PY - 1998

Y1 - 1998

N2 - Two related MRF models, an edge-preserving smoothing model followed by a modified standard regularisation, are presented for the adaptive binarisation of nonuniform images in the presence of noise. In particular, a computational model is developed for a modified standard regularisation method which calculates the adaptive threshold surface for noisy images. Since the modified standard regularisation depends only on the image data, and not its edge segments, it gives much better performance and can be applied to more classes of image than those methods that solely rely on edge segments. Experimental results demonstrate that the proposed method has the best performance over three other commonly used adaptive segmentation methods and is faster than previous interpolation-based thresholding techniques.

AB - Two related MRF models, an edge-preserving smoothing model followed by a modified standard regularisation, are presented for the adaptive binarisation of nonuniform images in the presence of noise. In particular, a computational model is developed for a modified standard regularisation method which calculates the adaptive threshold surface for noisy images. Since the modified standard regularisation depends only on the image data, and not its edge segments, it gives much better performance and can be applied to more classes of image than those methods that solely rely on edge segments. Experimental results demonstrate that the proposed method has the best performance over three other commonly used adaptive segmentation methods and is faster than previous interpolation-based thresholding techniques.

KW - Adaptive image binarisation

KW - Edge-preserving smoothing

KW - Markov random field models

KW - Standard regularisation

KW - Threshold surface

UR - http://www.scopus.com/inward/record.url?scp=0032183676&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032183676&partnerID=8YFLogxK

M3 - Chapter

AN - SCOPUS:0032183676

VL - 145

SP - 322

EP - 332

BT - IEE Proceedings: Vision, Image and Signal Processing

ER -