Seite - 171 - in Joint Austrian Computer Vision and Robotics Workshop 2020

lage attack (CA). For the patch-size chosen in the comparison, the imagequality isslightlysuperior for CA. GS and GS+LS as also proposed in literature areneithercompetitive in termsofmaintained image quality nor in terms of preventing the copy move at- tack detection capability. When considering the ease of application, LS is clearly preferable, as CA re- quires the generation of a keypoint-free dictionary and a vector-quantisation like patch selection pro- cess,whileLSonlyappliesa localGaussiansmooth- ing. Overall, LS turns out to be a highly attractive alternative to SIFT keypoint removal techniques ap- plied so far in literature. References [1] I. Amerini, M. Barni, R. Caldelli, and A. Costanzo. Counter-forensicsofsift-basedcopy-movedetection bymeansofkeypointclassification. EURASIPJour- nal on Image and Video Processing, 2013(1):18, 2013. [2] I. Amerini, M. Barni, R. Caldelli, and A. Costanzo. Removal and injection of keypoints for sift-based copy-move counter-forensics. EURASIP Journal on InformationSecurity, 2013(1):8, 2013. [3] I. Amerini, F. Battisti, R. Caldelli, M. Carli, and A. Costanzo. Exploiting perceptual quality is- sues in countering sift-based forensic methods. In 2014 IEEE International Conference on Acous- tics, Speech and Signal Processing (ICASSP), pages 2664–2668. IEEE,2014. [4] E. Ardizzone, A. Bruno, and G. Mazzola. Detect- ing multiple copies in tampered images. In 2010 IEEE International Conference on Image Process- ing, pages2117–2120. IEEE,2010. [5] E. Ardizzone, A. Bruno, and G. Mazzola. Copy– move forgery detection by matching triangles of keypoints. IEEE Transactions on Information ForensicsandSecurity, 10(10):2084–2094,2015. [6] V. Christlein, C. Riess, J. Jordan, C. Riess, and E. Angelopoulou. An evaluation of popular copy- move forgery detection approaches. IEEE Trans- actions on Information Forensics and Security, 7(6):1841–1854,2012. [7] A. Costanzo, I. Amerini, R. Caldelli, and M. Barni. Forensicanalysisof siftkeypoint removaland injec- tion. IEEE Transactions on Information Forensics andSecurity, 9(9):1450–1464,2014. [8] D.-T. Dang-Nguyen, C. Pasquini, V. Conotter, and G.Boato. Raise: arawimagesdataset fordigital im- age forensics. In Proceedings of the 6th ACM Mul- timedia Systems Conference, pages 219–224. ACM, 2015. [9] T.-T. Do, E. Kijak, T. Furon, and L. Amsaleg. De- luding image recognition in sift-based cbir systems. In Proceedings of the 2nd ACM workshop on Multi- media in forensics, security and intelligence, pages 7–12.ACM,2010. [10] C.-Y.Hsu,C.-S.Lu,andS.-C.Pei.Secureandrobust sift. In Proceedings of the 17th ACM International Conference on Multimedia, pages 637–640. ACM, 2009. [11] H.Huang,W.Guo,andY.Zhang.Detectionofcopy- move forgery in digital images using sift algorithm. In Pacific-Asia Workshop on Computational Intelli- genceandIndustrialApplication,2008.PACIIA’08., volume2, pages 272–276. IEEE, 2008. [12] D. G. Lowe. Distinctive image features from scale- invariant keypoints. International Journal of Com- puterVision, 60(2):91–110,November2004. [13] B. Mahdian and S. Saic. Detection of copy–move forgeryusingamethodbasedonblurmomentinvari- ants. Forensic Science International, 171(2-3):180– 189,2007. [14] MATLAB. version 8.3.0.532 (R2014a). The Math- Works Inc.,Natick, Massachusetts, 2014. [15] M.MuX. MeTriXMuXversion1.1. 2014. [16] J. Platt et al. Probabilistic outputs for support vec- tor machines and comparisons to regularized likeli- hood methods. Advances in Large Margin Classi- fiers, 10(3):61–74,1999. [17] M. Saleem. A key-point based robust algorithm for detecting cloning forgery. In IEEE International Conference on Control System, Computing and En- gineering (ICCSCE), volume 4, pages 2775–2779, 2014. [18] G. Schäfer and M. Stich. UCID - an uncompressed colour image database. Proc. SPICE. Storage and RetrievalMethodsandApplications forMultimedia, 11(1):472–480,2004. [19] H. Sencar and N. M. (Eds.). Digital Image Foren- sics: There is more to a picture than meets the eye. SpringerVerlag, 2012. [20] A. Vedaldi and B. Fulkerson. Vlfeat: An open and portable library of computer vision algorithms. In Proceedings of the 18th ACM International Confer- ence on Multimedia, MM ’10, pages 1469–1472, NewYork,NY, USA,2010.ACM. [21] M. Zandi, A. Mahmoudi-Aznaveh, and A. Man- souri. Adaptivematchingforcopy-moveforgeryde- tection. In 2014 IEEE International Workshop on Information Forensics and Security (WIFS), pages 119–124. IEEE,2014. [22] D. Zhang and G. Lu. Evaluation of similarity mea- surement for image retrieval. In International Con- ference on Neural Networks and Signal Processing, 2003. Proceedings of the 2003, volume 2, pages 928–931. IEEE,2003. 171