Intra-frame Adaptive Transform Size for Video Steganography in H.265/HEVC Bitstreams

Abstract

A video steganography method based on H.265/HEVC (High Efficiency Video Coding) video adaptive transform block size is proposed in this paper. With the adoption of new quadtree splitting structure in H.265/HEVC video compression standard, secret information could be embedded into carrier bitstreams by adaptive modifying the transform unit (TU) size during the procedures of searching best coding syntax elements set. Different from traditional video steganography methods, the proposed method could achieve at most modifying the transform unit splitting once while multiple secret information is embedded, which can achieve a high visual quality for digital carrier video. Moreover, the proposed method is only manipulated on the splitting decision process of TUs. Large volume quantity of TUs in compressed carrier video could guarantee a high embedding capacity. The experimental results have been proven the superiority of efficiency and performance about the proposed method. Moreover, the proposed method can be better applied to high-definition (HD) or higher video applications, and provides a practical tool for the protection of privacy data about digital video, as well as the contents of digital video itself.