Breaking the Digital Video Steganography

In this paper we provide a new digital video steganalysis algorithm. When frames of the cover video are embedded with secret messages, both number of connected component and number of holes will change dramatically, thus Euler number of the stego-frame will has pulsing increase. This proposed steganalysis algorithm is applied to stego-video, estimating steganographic algorithm for frames chosen and estimating steganographic capacity. Experimental results indicate that it’s much simpler and faster, more sensitive and robust, steganography rate as small as 0.0154% can be reliably detected. It’s effective and efficient to detect any format video, because this algorithm is only correlated with the connected components number and holes number of the frames.