Digital video steganalysis toward spread spectrum data hiding

作者：Zarmehi Nematollah; Akhaee Mohammad Ali^*

来源：IET Image Processing, 2016, 10(1): 1-8.

DOI：10.1049/iet-ipr.2014.1019

• data encapsulation
• steganography
• video signal processing
• feature extraction
• matrix algebra
• support vector machines
• digital video steganalysis
• spread spectrum data hiding
• steganalytic scheme
• digital video spread spectrum
• hidden message
• SS embedding rules
• residual matrix
• matrix extraction
• video frames
• support vector machine
• embedding process gain factor

In this study, the authors propose a steganalytic scheme for digital video spread spectrum (SS) data hiding. The proposed method estimates both the hidden message and the gain factor of the SS embedding rules. In this method, the cover frames are first estimated and are compared with the received video frames. Then, the residual matrix is computed and specific features are extracted from this matrix as well as the video frames and estimated frames. The support vector machine, then applies to the extracted features to classify the video as either clean or suspicious. If the video is declared suspicious, both the hidden message and the embedding process gain factor are estimated and consequently the original video is reconstructed. The simulation results confirm the success of the authors' proposed method in detecting the stego video, estimation of the hidden message and gain factor as well as reconstruction of the original video.

• 出版日期2016-1