Enhanced matching method for copy-move forgery detection by means of Zernike moments

Osamah M. Al-Qershi, Bee Ee Khoo

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

8 Citations (Scopus)

Abstract

Copy-move is one of the most popular and efficient operations to create image forgery. Many passive detection techniques have been proposed to detect such a forgery in digital images. The performance of the detection algorithms depends mainly on the features used for matching image blocks or keypoints and the matching method as well. Among the existing detection algorithms, those which employ Zernike moments as features provide remarkable detection accuracy. The robustness of Zernike moments comes from the fact that they are invariant to rotation and scaling. However, Zernike moments-based algorithms can be improved further by adopting more efficient matching methods. In this paper, we propose a new matching method in order to enhance the detection accuracy. Compared to the lexicographical sorting-based matching method, the proposed method improved the detection accuracy by 40 %.

Original languageEnglish
Title of host publicationDigital-Forensics and Watermarking
Subtitle of host publication13th International Workshop, IWDW 2014 Taipei, Taiwan, October 1–4, 2014 Revised Selected Papers
EditorsYun-Qing Shi, Hyoung Joong Kim, Fernando Pérez-González, Ching-Nung Yang
Place of PublicationCham Switzerland
PublisherSpringer
Pages485-497
Number of pages13
ISBN (Electronic)9783319193212
ISBN (Print)9783319193205
DOIs
Publication statusPublished - 2015
Externally publishedYes
EventInternational Workshop on Digital-Forensics and Watermarking 2014 - Taipei, Taiwan
Duration: 1 Oct 20144 Oct 2014
Conference number: 13th
https://web.archive.org/web/20140103103849/http://iwdw2014.ntcb.edu.tw/

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume9023
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceInternational Workshop on Digital-Forensics and Watermarking 2014
Abbreviated titleIWDW 2014
CountryTaiwan
CityTaipei
Period1/10/144/10/14
Internet address

Keywords

  • Detection Accuracy
  • Image Block
  • Zernike Moment
  • Discrete Cosine Transform Coefficient
  • Matching Block

Cite this