This paper proposes an original approach for the automatic detection of AI-generated images, using features derived from noise residuals artefacts. Contrary to most current research that leverages sophisticated deep learning models to further improve performance, this study highlights the distinct noise residual characteristics in deepfakes, facilitating the identification of AI-generative images. Our findings highlight some limitations of image models, which can be used for forensic analysis and for future AI-based text-to-image generative models. Broad numerical results on a large and diverse dataset show the interest of the identified features as well as the relevance of the present method.

This work explores the potential of synthetic media generated by the means of Artificial Intelligence (AI), sometimes referred to as Deepfakes, as a source of cover-objects for steganography. Deepfakes offer a vast and diverse pool of media, potentially improving steganographic security by leveraging cover-source mismatch, a challenge in steganalysis where training and testing data come from different sources. The present paper proposes an initial study on Deepfakes’ effectiveness in the field of steganography. More precisely, we propose an initial investigation to assess the impact of Deepfakes on image steganalysis performance in an operational environment. Using a wide range of image generation models and state-of-the-art methods in steganography and steganalysis, we show that Deepfakes can significantly exploit the cover-source mismatch problem but that mitigation solutions also exist. The empirical findings can inform future research on steganographic techniques that exploit cover-source mismatch for enhanced security.

The present paper deals with the cover-source mismatch (CSM) problem in operational steganalysis. It first investigates the distribution of the noise in natural images, and shows how this property can be used to build a fingerprint of the cover- source, to address the issue of source identification from a single image. In particular, fingerprints from different noise extraction techniques are studied. Results show that these fingerprints can be complementary. The method proposed in the present paper aggregates them in a unique forensic feature to build a more accurate source identification algorithm than when using steganalysis features, such as the discrete cosine transform residual (DCTR). Last, the paper exploits the proposed forensic tool to mitigate CSM via "atomistic steganalysis". Used together with steganalysis methods, experimental results highlight the superiority of our approach, as compared to other atomistic mitigation strategies. The relevancy of these results is further studied on out- of-camera images coming from Flickr and the ALASKA dataset. We show that for some devices, our approach gives results superior to the omniscient scenario.

This paper deals with the Cover-Source Mismatch (CSM) problem faced in operational steganalysis. Based on a multivariate Gaussian model of the distribution of the noise contained in natural images, it provides proxies for the two important empirical measures of CSM: intrinsic difficulty and regret. The former can be modeled with the determinant of the covariance matrix of the noise present in an image. The latter can be predicted with a modified Kullback-Leibler divergence between the distribution of the noises of images coming from different cover-sources. We first recall the reasoning behind the multivariate Gaussian model of the noise, and detail how to compute the statistic of the distribution of the noise. Then, our proposed models are compared to empirical data with a specifically designed cover-source generation process. For both quantities, very high correlation coefficients between the model and the observations are obtained. Finally, realistic cover-sources are used to further illustrate the relevance of our model.

Operational steganalysis contends with a major problem referred to as the cover-source mismatch (CSM), which is essentially a difference in distribution caused by different parameters and settings over training and test data. Despite it being of fundamental importance in an operational context, the CSM problem is often overlooked in the literature. With the goal to increase the visibility of this problem and attract the interest of the community, the present paper proposes a systematic review of the literature. It summarizes gathered knowledge and major open questions over the last 20 years of active research on CSM: terminology, methods of measurement, known causes, and mitigation strategies. Over 100 papers exploring, mitigating, assessing, or discussing steganalysis under train-test mismatch were collected by sampling scholar databases, and tracing references, cited and generated. For image steganalysis, the literature provided enough evidence to quantify the impact of causes, and the effectiveness of mitigation strategies.