A Source Video Identification Algorithm Based on Motion Vectors

With the easy availability of DV (digital video) and associated software to manipulate videos, ascertaining the integrity and authenticity of digital videos has become an urgent and critical issue in today’s digital community. In this paper, a new source video system identification algorithm is proposed based on the motion vector information in the encoded stream; it takes full advantage of the various characteristics in the motion estimation algorithm in different video compression systems, and combines a k-nearest neighbor (k-NN) classifier to build a complete video system identification scheme. The experiments show this proposed algorithm can effectively identify video streams which come from a number of video coding systems.

[1]  Beatrice Lazzerini,et al.  K-NN algorithm based on neural similarity , 2002, Proceedings 2002 IEEE International Conference on Artificial Intelligence Systems (ICAIS 2002).

[2]  Edmund Y Lam,et al.  Automatic source camera identification using the intrinsic lens radial distortion. , 2006, Optics express.

[3]  Miroslav Goljan,et al.  Digital camera identification from sensor pattern noise , 2006, IEEE Transactions on Information Forensics and Security.

[4]  Nasir D. Memon,et al.  Source camera identification based on CFA interpolation , 2005, IEEE International Conference on Image Processing 2005.

[5]  Mo Chen,et al.  Digital imaging sensor identification (further study) , 2007, Electronic Imaging.

[6]  Hany Farid,et al.  Digital doctoring: how to tell the real from the fake , 2006 .

[7]  Jan P. Allebach,et al.  A survey of forensic characterization methods for physical devices , 2006, Digit. Investig..

[8]  Mohan S. Kankanhalli,et al.  A Survey on Digital Camera Image Forensic Methods , 2007, 2007 IEEE International Conference on Multimedia and Expo.