Transfer subspace learning based on structure preservation for JPEG image mismatched steganalysis

Yang, Liran; Men, Min; Xue, Yiming; Wen, Juan; Zhong, Ping

doi:10.1016/j.image.2020.116052

Cited by 10 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The previous steganalysis methods habitually assumed that the training data set and the test set have the same probability distribution (for example, from the same data set). But in practice, this constraining assumption is difficult to be fulfilled [8,15,18]. For example, the model we trained in the source domain (such as Twitter) may encounter the text from the movie review domain or news domain at the test time, which, unfortunately, have almost no labeled data.…”

Section: Domain Mismatch In Text Steganalysismentioning

confidence: 99%

“…Recent research attempts to train a multi-classifier to recognize Image Processing Pipeline (IPP) first, and then select the training set corresponding to the IPP to train a steganalysis classifier [2]. There are also studies projecting training and testing data into a common subspace, learning domain-invariant representations, and directly applying classifiers trained on source-domain data to target instances [8,15]. These methods are effective for images, but for generative text steganography trained on large-scale corpora, the distribution of steganographic text is more complex.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Domain Adaptational Text Steganalysis Based on Transductive Learning

Xue

Yang

Deng

et al. 2022

Proceedings of the 2022 ACM Workshop on Information Hiding and Multimedia Security

Self Cite

View full text Add to dashboard Cite

Traditional text steganalysis methods rely on a large amount of labeled data. At the same time, the test data should be independent and identically distributed with the training data. However, in practice, a large number of text types make it difficult to satisfy the i.i.d condition between the training set and the test set, which leads to the problem of domain mismatch and significantly reduces the detection performance. In this paper, we draw on the ideas of domain adaptation and transductive learning to design a novel text steganalysis method. In this method, we design a distributed adaptation layer and adopt three loss functions to achieve domain adaptation, so that the model can learn the domaininvariant text features. The experimental results show that the method has better steganalysis performance in the case of domain mismatch. CCS CONCEPTS• Security and privacy → Human and societal aspects of security and privacy; • Computing methodologies → Natural language processing.

show abstract

Section: Domain Mismatch In Text Steganalysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Domain Adaptational Text Steganalysis Based on Transductive Learning

Xue

Yang

Deng

et al. 2022

Proceedings of the 2022 ACM Workshop on Information Hiding and Multimedia Security

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is relevant, contrary to the atomistic approach, when dealing with unseen cover-sources. This approach is promising but can suffer from the double impact of steganography and CSM on current state-of-the-art features [25].…”

Section: Introductionmentioning

confidence: 99%

Statistical Correlation as a Forensic Feature to Mitigate the Cover-Source Mismatch

Mallet,

Bas,

Cogranne

2024

Proceedings of the 2024 ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

“…Bessaoudi et al [3] proposed a linear multi-perspective subspace learning method for face kinship verification in the wild. Yang et al [4] proposed a transfer subspace learning method by preserving the image structure, which is used to analyze encrypted images. Qin et al [5] developed a structured subspace learning method that induces symmetric non-negative matrix factorization to learn similar subspaces and latent subspaces.…”

Section: Introductionmentioning

confidence: 99%

An Improved PixelHop Framework and its Application in Rolling Bearing Fault Diagnosis

Wan¹,

Zhou

Gong

et al. 2021

IEEE Access

View full text Add to dashboard Cite

The PixelHop framework based on successive subspace learning (SSL) has been widely used in signal processing and computer vision, which can effectively improve the classification accuracy in high spatial resolution scenes through successive subspace growth. To solve the problems of insufficient feature extraction and dependence on prior knowledge in the PixelHop framework, an improved PixelHop (I-PixelHop) framework is proposed. On the basis of PixelHop framework, I-PixelHop has made the following improvements. 1) I-PixelHop fully extracts the continuous features in one-dimensional sequence data through the improved neighborhood expansion, which can provide a richer feature set. 2) The improved label-assisted regression (ILAG) unit uses the Bi-K-Means clustering algorithm to enable more correct clustering of similar samples, and it adopts the cross-entropy threshold method to alleviate the negative effects caused by the improper setting of the number of pseudo-classes. 3) The high-dimensional features are fully reused by adopting the pseudo dense connection structure to obtain a better feature set. Moreover, the proposed I-PixelHop framework is applied to the rolling bearing fault diagnosis. A series of experiments are carried out to verify the effectiveness of the proposed I-PixelHop. The experimental results show that the fault diagnosis accuracy of I-PixelHop can reach 98.91% and 98.74% on the two different rolling bearing fault datasets, and it also has satisfactory anti-noise ability, faster training speed, and smaller model size.

show abstract

Transfer subspace learning based on structure preservation for JPEG image mismatched steganalysis

Cited by 10 publications

References 39 publications

Domain Adaptational Text Steganalysis Based on Transductive Learning

Domain Adaptational Text Steganalysis Based on Transductive Learning

Statistical Correlation as a Forensic Feature to Mitigate the Cover-Source Mismatch

An Improved PixelHop Framework and its Application in Rolling Bearing Fault Diagnosis

Contact Info

Product

Resources

About