Development of low-complexity matrix embedding with an efficient iterative strategy

Hsi Yuan Chang, Jyun Jie Wang, Chi Yuan Lin, Chin-Hsing Chen

Research output: Contribution to journalArticle

Abstract

A novel suboptimal embedding algorithm for binary logos based on a weight approach embedding (WAE) is proposed. An optimal embedding algorithm, developed on the basis of the maximal likelihood algorithm, is aimed at locating the coset leader as an approach to the minimum embedding distortion. By contrast, there is no need to locate the coset leader; instead, a target vector is required. The corresponding weight of the target vector is close to that of the coset leader; the target vector’s weight is discovered in an efficiently iterative manner. In case of a highest operating complexity, in contrast to that of the optimal maximal likelihood (ML) algorithm, the operating complexity of the suboptimal WAE is linearly proportional to the number of code dimensions, because a full search (which would be necessary in an ML algorithm) is not required. The lower embedding efficiency of the proposed algorithm is superior to those of algorithms proposed in previous works, when the logo message length is small.

Original languageEnglish
Pages (from-to)707-713
Number of pages7
JournalICIC Express Letters, Part B: Applications
Volume9
Issue number7
DOIs
Publication statusPublished - 2018 Jul 1

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

@article{cd353dd5045d4f248f3d19a0baf6dc33,
title = "Development of low-complexity matrix embedding with an efficient iterative strategy",
abstract = "A novel suboptimal embedding algorithm for binary logos based on a weight approach embedding (WAE) is proposed. An optimal embedding algorithm, developed on the basis of the maximal likelihood algorithm, is aimed at locating the coset leader as an approach to the minimum embedding distortion. By contrast, there is no need to locate the coset leader; instead, a target vector is required. The corresponding weight of the target vector is close to that of the coset leader; the target vector’s weight is discovered in an efficiently iterative manner. In case of a highest operating complexity, in contrast to that of the optimal maximal likelihood (ML) algorithm, the operating complexity of the suboptimal WAE is linearly proportional to the number of code dimensions, because a full search (which would be necessary in an ML algorithm) is not required. The lower embedding efficiency of the proposed algorithm is superior to those of algorithms proposed in previous works, when the logo message length is small.",
author = "Chang, {Hsi Yuan} and Wang, {Jyun Jie} and Lin, {Chi Yuan} and Chin-Hsing Chen",
year = "2018",
month = "7",
day = "1",
doi = "10.24507/icicelb.09.07.707",
language = "English",
volume = "9",
pages = "707--713",
journal = "ICIC Express Letters, Part B: Applications",
issn = "2185-2766",
publisher = "ICIC Express Letters Office",
number = "7",

}

Development of low-complexity matrix embedding with an efficient iterative strategy. / Chang, Hsi Yuan; Wang, Jyun Jie; Lin, Chi Yuan; Chen, Chin-Hsing.

In: ICIC Express Letters, Part B: Applications, Vol. 9, No. 7, 01.07.2018, p. 707-713.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of low-complexity matrix embedding with an efficient iterative strategy

AU - Chang, Hsi Yuan

AU - Wang, Jyun Jie

AU - Lin, Chi Yuan

AU - Chen, Chin-Hsing

PY - 2018/7/1

Y1 - 2018/7/1

N2 - A novel suboptimal embedding algorithm for binary logos based on a weight approach embedding (WAE) is proposed. An optimal embedding algorithm, developed on the basis of the maximal likelihood algorithm, is aimed at locating the coset leader as an approach to the minimum embedding distortion. By contrast, there is no need to locate the coset leader; instead, a target vector is required. The corresponding weight of the target vector is close to that of the coset leader; the target vector’s weight is discovered in an efficiently iterative manner. In case of a highest operating complexity, in contrast to that of the optimal maximal likelihood (ML) algorithm, the operating complexity of the suboptimal WAE is linearly proportional to the number of code dimensions, because a full search (which would be necessary in an ML algorithm) is not required. The lower embedding efficiency of the proposed algorithm is superior to those of algorithms proposed in previous works, when the logo message length is small.

AB - A novel suboptimal embedding algorithm for binary logos based on a weight approach embedding (WAE) is proposed. An optimal embedding algorithm, developed on the basis of the maximal likelihood algorithm, is aimed at locating the coset leader as an approach to the minimum embedding distortion. By contrast, there is no need to locate the coset leader; instead, a target vector is required. The corresponding weight of the target vector is close to that of the coset leader; the target vector’s weight is discovered in an efficiently iterative manner. In case of a highest operating complexity, in contrast to that of the optimal maximal likelihood (ML) algorithm, the operating complexity of the suboptimal WAE is linearly proportional to the number of code dimensions, because a full search (which would be necessary in an ML algorithm) is not required. The lower embedding efficiency of the proposed algorithm is superior to those of algorithms proposed in previous works, when the logo message length is small.

UR - http://www.scopus.com/inward/record.url?scp=85047979192&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047979192&partnerID=8YFLogxK

U2 - 10.24507/icicelb.09.07.707

DO - 10.24507/icicelb.09.07.707

M3 - Article

VL - 9

SP - 707

EP - 713

JO - ICIC Express Letters, Part B: Applications

JF - ICIC Express Letters, Part B: Applications

SN - 2185-2766

IS - 7

ER -