Detection of copy–move forgery using discrete analytical Fourier–Mellin transform

被引:0
作者
Junliu Zhong
Yanfen Gan
机构
[1] Guangdong Mechanical and Electrical College,School of Information Engineering
[2] Guangdong University of Foreign Studies South China Business College,Department of Information Science and Technology
来源
Nonlinear Dynamics | 2016年 / 84卷
关键词
Copy–move forgery; Post-processed forgery operations ; Geometric distortions; Discrete analytical Fourier–Mellin transform; A disk template;
D O I
暂无
中图分类号
学科分类号
摘要
In recent years, digital image processing has become commonplace with growing powerful and available image editing software. People without any professional technique can also manipulate and forge digital images easily. One of the most popular manners of digital image forgeries is the copy–move image forgery. Extensive researches in detecting copy–move forgery have made a deal of achievements, but most presented methods based on these researches have been only focus on some simple composite forgeries and not able to detect different types of post-processed forgeries. In this paper, we aim to deal with the post-processed forgery operations and scenarios, mainly geometric distortion. We introduce analytical Fourier–Mellin transform (AFMT) and focus on its discretization. We propose discrete analytical Fourier–Mellin transform (DAFMT). We also pay attention to high performance of DAFMT in detecting the copy–move image forgeries. Due to the AFMT described in polar coordinate, so we need to convert coordinate system from polar to Cartesian coordinates. To be computed conveniently, we define an auxiliary disk template to accomplish this conversion. We devote to the use of our proposed DAFMT in detection of image forgeries. A great deal of researches and experiments show that the proposed DAFMT can effectively resist translation, rotation, scaling, and added Gaussian noise operations. Compared with other relevant up-to-date methods, experiments also prove that DAFMT has made a progress in detecting and identifying the forgery images which are suffered from geometric distortion operations.
引用
收藏
页码:189 / 202
页数:13
相关论文
共 22 条
  • [1] Farid H(2009)A survey of image forgery detection IEEE Signal Process. Mag. 26 16-25
  • [2] Popescu AC(2005)Exposing digital forgeries by detecting traces of resampling IEEE Trans. Signal Process. 53 758-767
  • [3] Hu MK(1962)Visual pattern recognition by moment invariants IEEE Trans. Info. Theory 8 179-187
  • [4] Liao S(1998)On the accuracy of Zernike moments for image analysis IEEE Trans. Pattern Anal. Mach. Intell. 20 1358-1364
  • [5] Pawlak M(2010)Two-dimensional polar harmonic transforms for invariant image representation IEEE Trans. Pattern Anal. Mach. Intell. 32 1260-1270
  • [6] Yap PT(2011)A sift-based forensic method for copy-move attack detection and transformation recovery IEEE Trans. Info. Forensics Secur. 6 1099-1110
  • [7] Jiang XD(2013)Rotation invariant localization of duplicated image regions based on Zernike moments IEEE Trans. Info. Forensics Secur. 8 1355-1370
  • [8] Kot AC(2013)Quaternion polar harmonic transforms for color images IEEE Signal Process. Lett. 20 803-807
  • [9] Amerini I(2007)Robust detection of region duplication forgery in digital image Chin. J. Comput. 30 1998-2007
  • [10] Ballan L(2012)Exposing post processed copy-paste forgeries through transform-invariant features IEEE Trans. Info. Forensics Secur. 7 1018-1028