(Qingdao university,Qingdao,266071,China.Xu Chang-ying)
Abstract:Copy-move is a common forgery in the research field of digital image forensics.The existing algorithms on copy-move detecting are quite computationally expensive and low operating efficiency for lexicographically order.In order to overcome the drawback,an autocorrelation detection methods based on the discrete wavelet transform is proposed in this paper.In this method,low-frequency sub-band wavelet decomposition which is best approximation of the image is extracted as a feature vector and calculate their Pearson correlation coefficient to detect the similar blocks.The experiments demonstrate that the proposed approach can greatly reduce the dimension of feature vector and accordingly increased detection efficiency of the matching block. Keywords: digital image; copy-move forgery; discrete wavelet transform; passive image forensics
CLC Number:TP391 Document Code: A Article ID:
References
[1]Zhao Jun-hong.Survey on Techniques of Bland Digital Image Forensics[J].Computer Engineering and Technology,2011,32(8):55-59.
[2]Wang Rui,Fang Yong.Blind Detection of Copy-Move Forgery in Digital Images Based on Moment Invariants[J].Journal of Image and Graphics,2008,(10):1938-1941.
[3]Fang Jun-Li.Natural Image of Copy and Paste and Fuzzy Operation to Tamper with Blind Forensics[D].Master Degree Theses of Master of Chinese Full Test Database,2009.
[4]Fridrich J,Soukal D,Luka J.Detection of Copy-move Forgery in Digital Images[C].Proc Digital Forensic Research Workshop,Cleveland,OH,August 2003.
[5]Wu Qiong.Li Guo-Hui.The Image Copy Forged Area Detection Based on Wavelet and Singular Value Decomposition[J]. Min-Micro,2008,29(4):730-733.
[6]Wei Wei-Ming,Passive Authentication of Copy-Move Forgery in Digital Images[J].Journal of Southeast University(Natural Science Edition),2007,37(1):58-60.
[7]He Dong-Jian.Digital Image Processing[M]Xi’an:Xi’an University Press,2003.
[8]Li Feng,Liu Yun-song.A Robust Detection Algorithm for Image Region-Duplication Forgery[J].Microcomputer information,2011,27(11):9-11.
[9]Zhou Lin-na.Digital Image Forensics[M]Beijing:University of Post Press,2008.
Author:Xu Chang-ying, gender female,postgraduate student,from Qingdao university, the direction of research is passive image forensics.
Wang Ying,gender female,doctor,master Tutor.