A Modified Fourier-Mellin Approach for Source Device Identification on Stabilized Videos
This work addresses a practical bottleneck in multimedia forensics for law enforcement or security applications, but it is incremental as it builds on existing Fourier-Mellin approaches.
The paper tackled the problem of aligning noise patterns for source device identification in stabilized videos by proposing a frequency-domain method to search for scaling and rotation parameters, achieving promising results on a real-world dataset.
To decide whether a digital video has been captured by a given device, multimedia forensic tools usually exploit characteristic noise traces left by the camera sensor on the acquired frames. This analysis requires that the noise pattern characterizing the camera and the noise pattern extracted from video frames under analysis are geometrically aligned. However, in many practical scenarios this does not occur, thus a re-alignment or synchronization has to be performed. Current solutions often require time consuming search of the realignment transformation parameters. In this paper, we propose to overcome this limitation by searching scaling and rotation parameters in the frequency domain. The proposed algorithm tested on real videos from a well-known state-of-the-art dataset shows promising results.