Selection-channel-aware rich model for Steganalysis of digital images

By Tomáš Denemark, Vahid Sedighi, Vojtech Holub, Rémi Cogranne, Jessica J. Fridrich. International Workshop on Information Forensics and Security, 2014

366 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/3a3b1bfb748a2fb77f49f7a4e519ef947219d643

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Bin Li
2. Gongshen Liu
3. Feng Zhu
4. Ru Zhang
5. Jianyi Liu
6. Mo Chen
7. Mehdi Boroumand
8. Yi Zhang

Citations in the corpus

• Deep residual learning for image steganalysis (2018, methodology): “To demonstrate the effectiveness of the DRN for image steganalysis, we compare its performances with the SRM [8] and maxSRMd2 [6].”
• Yedroudj-Net: An Efficient CNN for Spatial Steganalysis (2018, neutral): “…the gain obtained in other image processing tasks using Deep Learning meth-ods [10], the steganalysis results were not ”10% better” compared to the classical approaches that use an Ensemble Classifier [11] with a Rich Model [12, 13] or a Rich Model with a Selection-Channel Awareness [14, 15].”
• Depth-Wise Separable Convolutions and Multi-Level Pooling for an Efficient Spatial CNN-Based Steganalysis (2020, methodology): “The existing steganalysis methods are usually classified into specific steganalysis algorithms [5], [48], [49] and universal steganalysis algorithms [13], [15], [44].”
• Steganalysis of Adaptive JPEG Steganography Using 2D Gabor Filters (2015, methodology): “In literature [3], the 2D Gabor filter has been used for the distortion definition of adaptive JPEG steganography, and the better anti-detection ability is achieved than db8 wavelet.”
• A New Payload Partition Strategy in Color Image Steganography (2020, background): “This is a result that was inspired by game-theoretical analyses of steganography [35], [36] and can also be seen in WOW being insecure against the new stegana-lytic feature sets (e.g., maxSRMd2 [20]).”
• Deep Residual Network for Steganalysis of Digital Images (2019, methodology): “the old detection paradigm based on the ensemble classifier and rich models, we steganalyzed all spatial-domain embedding algorithms with the maxSRMd2 [17] features non-linearly normalized using random conditioning (RC) [8].”