Variational image binarization and its multi-scale realizations

Chong Sze TONG; Yongping Zhang; Nanning Zheng

doi:10.1007/s10851-005-6466-x

Variational image binarization and its multi-scale realizations

Chong Sze TONG^*, Yongping Zhang, Nanning Zheng

^*Corresponding author for this work

Department of Mathematics

Research output: Contribution to journal › Review article › peer-review

12 Citations (Scopus)

Abstract

A variational approach for image binarization is discussed in this paper. The approach is based on the interpolation of surface. This interpolation is computed using edge points as interpolating points and minimizing an energy functional which interpolates a smooth threshold surface. A globally convergent Sequential Relaxation Algorithm (SRA) is proposed for solving the optimization problem. Moreover, our algorithm is also formulated in a multi-scale framework. The performance of our method is demonstrated on a variety of real and synthetic images and compared with traditional techniques. Examples show that our method gives promising results.

Original language	English
Pages (from-to)	185-198
Number of pages	14
Journal	Journal of Mathematical Imaging and Vision
Volume	23
Issue number	2
DOIs	https://doi.org/10.1007/s10851-005-6466-x
Publication status	Published - Sept 2005

Scopus Subject Areas

Statistics and Probability
Modelling and Simulation
Condensed Matter Physics
Computer Vision and Pattern Recognition
Geometry and Topology
Applied Mathematics

User-Defined Keywords

Global convergence
Image segmentation
Multi-scale image processing
Sequential relaxation algorithm
Variational approach

Access to Document

10.1007/s10851-005-6466-x

Cite this

@article{914ec5b7e60640e797f6e3cc6f29b68c,

title = "Variational image binarization and its multi-scale realizations",

abstract = "A variational approach for image binarization is discussed in this paper. The approach is based on the interpolation of surface. This interpolation is computed using edge points as interpolating points and minimizing an energy functional which interpolates a smooth threshold surface. A globally convergent Sequential Relaxation Algorithm (SRA) is proposed for solving the optimization problem. Moreover, our algorithm is also formulated in a multi-scale framework. The performance of our method is demonstrated on a variety of real and synthetic images and compared with traditional techniques. Examples show that our method gives promising results.",

keywords = "Global convergence, Image segmentation, Multi-scale image processing, Sequential relaxation algorithm, Variational approach",

author = "TONG, {Chong Sze} and Yongping Zhang and Nanning Zheng",

note = "Funding Information: ∗This research is partially supported by HKBU Faculty Research Grant FRG/02-03/II-04 and NSF of China Grant. †Now at Bioengineering Institute, University of Auckland, Level 6, 70 Symonds Street, Auckland, New Zealand.",

year = "2005",

month = sep,

doi = "10.1007/s10851-005-6466-x",

language = "English",

volume = "23",

pages = "185--198",

journal = "Journal of Mathematical Imaging and Vision",

issn = "0924-9907",

publisher = "Springer Netherlands",

number = "2",

}

TY - JOUR

T1 - Variational image binarization and its multi-scale realizations

AU - TONG, Chong Sze

AU - Zhang, Yongping

AU - Zheng, Nanning

N1 - Funding Information: ∗This research is partially supported by HKBU Faculty Research Grant FRG/02-03/II-04 and NSF of China Grant. †Now at Bioengineering Institute, University of Auckland, Level 6, 70 Symonds Street, Auckland, New Zealand.

PY - 2005/9

Y1 - 2005/9

N2 - A variational approach for image binarization is discussed in this paper. The approach is based on the interpolation of surface. This interpolation is computed using edge points as interpolating points and minimizing an energy functional which interpolates a smooth threshold surface. A globally convergent Sequential Relaxation Algorithm (SRA) is proposed for solving the optimization problem. Moreover, our algorithm is also formulated in a multi-scale framework. The performance of our method is demonstrated on a variety of real and synthetic images and compared with traditional techniques. Examples show that our method gives promising results.

AB - A variational approach for image binarization is discussed in this paper. The approach is based on the interpolation of surface. This interpolation is computed using edge points as interpolating points and minimizing an energy functional which interpolates a smooth threshold surface. A globally convergent Sequential Relaxation Algorithm (SRA) is proposed for solving the optimization problem. Moreover, our algorithm is also formulated in a multi-scale framework. The performance of our method is demonstrated on a variety of real and synthetic images and compared with traditional techniques. Examples show that our method gives promising results.

KW - Global convergence

KW - Image segmentation

KW - Multi-scale image processing

KW - Sequential relaxation algorithm

KW - Variational approach

UR - http://www.scopus.com/inward/record.url?scp=23944498708&partnerID=8YFLogxK

U2 - 10.1007/s10851-005-6466-x

DO - 10.1007/s10851-005-6466-x

M3 - Review article

AN - SCOPUS:23944498708

SN - 0924-9907

VL - 23

SP - 185

EP - 198

JO - Journal of Mathematical Imaging and Vision

JF - Journal of Mathematical Imaging and Vision

IS - 2

ER -

Variational image binarization and its multi-scale realizations

Abstract

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this