New design paradigm of distortion cost function for efficient JPEG steganography

• Published in: Signal processing 2022-01, Vol.190, p.108319, Article 108319
• Main Authors: Su, Wenkang, Ni, Jiangqun, Hu, Xianglei, Huang, Jiwu
• Format: Article
• Language: English
• Subjects: Distortion cost function; Domain transformation; Exponential model; Information hiding; JPEG steganography
• ISSN: 0165-1684; 1872-7557
• DOI: 10.1016/j.sigpro.2021.108319

•A novel design paradigm of distortion cost function for efficient JPEG steganography is proposed, which incorporates the statistics of both the spatial and DCT domains by a generalized Distortion Cost Domain Transformation (DCDT) function.•The proposed JPEG steganographic distortion cost function is represented in a more general exponential model, aiming to flexibly allocate the embedding data.•The proposed scheme can be easily extended to the mutually dependent embedding with the help of the proposed DCDT function.•Experimental results show that the proposed scheme can outperform other SOTA JPEG steganographic schemes, e.g., UERD, J-UNIWARD and GUED, and rival the BET with an order of magnitude lower computational complexity.•The proposed scheme can not only improve the performance against JPEG phase-aware steganalysis, but also broaden the applications of existing image steganographic schemes in spatial domain. Recently, with the introduction of JPEG (Joint Photographic Experts Group) phase-aware steganalysis features, e.g., GFR, the design of JPEG steganographic distortion cost function turns to maintain not only the statistical undetectability in DCT (Discrete Cosine Transform) domain but also in spatial domain. To tackle this issue, this paper presents a novel paradigm for the design of JPEG steganographic distortion cost function, which calculates the distortion cost via a generalized Distortion Cost Domain Transformation (DCDT) function. The proposed function comprises the decompressed pixel block embedding changes and their corresponding embedding distortion costs for unit change, where the pixel embedding distortion costs are represented in a more general exponential model, aiming to flexibly allocate the embedding data. In this way, the JPEG steganography could be formulated as the optimization problem of minimizing the overall distortion cost in its decompressed spatial domain, which is equivalent to maximizing its statistical undetectability against JPEG phase-aware steganalysis features. Experimental results show that the proposed DCDT equipped with HiLL (a spatial steganographic distortion cost function) is superior to other state-of-the-art JPEG steganographic schemes, e.g., UERD, J-UNIWARD, and GUED in resisting the detection of JPEG phase-aware feature-based steganalyzers GFR and SCA-GFR, and rivals BET-HiLL with one order of magnitude lower computational complexity, along with the possibility of being further improved by considering the