An automatic video-object based steganographic system for multi-use message hiding using wavelet transform

摘要

An automatic video-object based steganographic system is proposed in this paper for multiuse message hiding. Initially video objects from different images are selected and for each video object a message is embedded into its most significant wavelet coefficients to provide invisibility and resistance against lossy transmission or other distortion. The architecture consists of three modules. In the first module initial messages are enciphered using one or more encryption algorithms. The enciphered messages are imprinted onto white-background images to construct the message-images to be hidden. In the second module each of the cover video objects is decomposed into two levels with seven subbands, using the DWT. Next Qualified Significant Wavelet Trees (QSWTs), which are paths of significant wavelet coefficients, are estimated for the highest energy pair of subbands of each video object. During the third module each message-image is redundantly embedded to the coefficients of the best QSWTs of a video object and the IDWT is applied to provide the stego-object. In the last stage stego-objects are combined to provide the final content to be transmitted. Experimental results under various loss rates indicate the robustness and efficiency of the proposed steganographk system.