Steganalysis of Adaptive JPEG Steganography Using 2D Gabor Filters

Song, Xiaocheng; Liu, Fenlin; Yang, Ching-Nung; Luo, Xiangyang; Zhang, Yi

doi:10.1145/2756601.2756608

Cited by 261 publications

(150 citation statements)

References 26 publications

Supporting

Mentioning

141

Contrasting

Unclassified

Order By: Relevance

“…us, the proposed scenario is for a known algorithm and known embedding rate a ack. To analyze the proposed approach, we have used the well known Rich Models (RM) framework [7] for the spatial domain and Gabor Filter Residuals (GFR) [22] for the transformed domain, with Ensemble Classi ers (EC) [13].…”

Section: Resultsmentioning

confidence: 99%

Detection of Classifier Inconsistencies in Image Steganalysis

Lerch-Hostalot

Megías

2019

Proceedings of the ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

In this paper, a methodology to detect inconsistencies in classi cation-based image steganalysis is presented. e proposed approach uses two classi ers: the usual one, trained with a set formed by cover and stego images, and a second classi er trained with the set obtained a er embedding additional random messages into the original training set. When the decisions of these two classi ers are not consistent, we know that the prediction is not reliable. e number of inconsistencies in the predictions of a testing set may indicate that the classi er is not performing correctly in the testing scenario. is occurs, for example, in case of cover source mismatch, or when we are trying to detect a steganographic method that the classi er is no capable of modelling accurately. We also show how the number of inconsistencies can be used to predict the reliability of the classi er (classi cation errors).

show abstract

Section: Resultsmentioning

confidence: 99%

Detection of Classifier Inconsistencies in Image Steganalysis

Lerch-Hostalot

Megías

2019

Proceedings of the ACM Workshop on Information Hiding and Multimedia Security

View full text Add to dashboard Cite

show abstract

“…The CNN model trained at the 100000-th iteration is used as the steganalyzer. • GFR steganalyzer [20], denoted as φ ′ . It is based on 17000 histogram features generated with Gabor filters and an FLD ensemble classifier [21].…”

Section: A Settingsmentioning

confidence: 99%

CNN-Based Adversarial Embedding for Image Steganography

Tang

Tan

et al. 2019

IEEE Trans.Inform.Forensic Secur.

239

132

View full text Add to dashboard Cite

Historically, steganographic schemes were designed in a way to preserve image statistics or steganalytic features. Since most of the state-of-the-art steganalytic methods employ a machine learning (ML) based classifier, it is reasonable to consider countering steganalysis by trying to fool the ML classifiers. However, simply applying perturbations on stego images as adversarial examples may lead to the failure of data extraction and introduce unexpected artefacts detectable by other classifiers. In this paper, we present a steganographic scheme with a novel operation called adversarial embedding, which achieves the goal of hiding a stego message while at the same time fooling a convolutional neural network (CNN) based steganalyzer. The proposed method works under the conventional framework of distortion minimization. Adversarial embedding is achieved by adjusting the costs of image element modifications according to the gradients backpropagated from the CNN classifier targeted by the attack. Therefore, modification direction has a higher probability to be the same as the sign of the gradient. In this way, the so called adversarial stego images are generated. Experiments demonstrate that the proposed steganographic scheme is secure against the targeted adversary-unaware steganalyzer. In addition, it deteriorates the performance of other adversary-aware steganalyzers opening the way to a new class of modern steganographic schemes capable to overcome powerful CNN-based steganalysis. Index TermsSteganography, steganalysis, adversarial machine learning. I. INTRODUCTIONImage steganography is the art and science of concealing covert information within images. It is usually achieved by modifying image elements, such as pixels or DCT coefficients. On the other side of the game, steganalysis aims to reveal the presence of secret information by detecting whether there are abnormal artefacts left by data embedding.The developing history of steganography and steganalysis is rich of interesting stories, as they compete with each other and they benefit and evolve from the competition [1]. The earliest steganographic method was implemented by substituting the least significant bits of image elements with message bits. The stego artefacts introduced by this method can be effectively detected by Chi-squared attack [2], or steganalytic features based on first-order statistics [3]. In this initial phase of the competition, statistical hypothesis testing or a simple linear classifier such as FLD (Fisher Linear Discriminant) could serve the need of steganalysis.The first-order statistics can be restored after data embedding, as done in [4]. The abnormal artefacts in the first-order statistics can also be avoided as in [5], [6]. As a consequence, more powerful steganalytic features based on the second-order statistics [7], [8] were proposed. In this period, advanced machine learning (ML) tools, such as SVM (Support Vector Machine), were operated on high-dimensional features (where the dimension is typically several hundreds). These met...

show abstract

“…We choose as Single Image Detector (SID), which is referred in this paper as the function f , the feature-based Quantitative steganalysis algorithm proposed in [12], applied on the 17,000-dimensional JPEG domain Rich Model -the Gabor features residuals (GFR) [21]. The Quantitative steganalysis algorithm is a machine learning regression framework that assembles, via the process of gradient boosting, a large number of simpler base learners built on random subspaces of the original high-dimensional feature space.…”

Section: B Single Image Detector (Sid)mentioning

confidence: 99%

“…Also note that each time a SID is used (in the first or second test, during the learning and also during the test, on a cover or on a stego whatever the payload size), for a given input image, a feature vector Gabor Features Residuals (GFR) [21] of dimension 17 000 is first extracted. This feature vector is then cleaned from NaN values (it occurs when the feature values are constant over images) and from constant values, to obtain a 16 750-dimensional feature vectors.…”

Section: A Data Preparationmentioning

confidence: 99%

Pooled Steganalysis in JPEG: how to deal with the spreading strategy?

Zakaria

Chaumont

Subsol

2019

2019 IEEE International Workshop on Information Forensics and Security (WIFS)

View full text Add to dashboard Cite

In image pooled steganalysis, a steganalyst, Eve, aims to detect if a set of images sent by a steganographer, Alice, to a receiver, Bob, contains a hidden message. We can reasonably assess that the steganalyst does not know the strategy used to spread the payload across images. To the best of our knowledge, in this case, the most appropriate solution for pooled steganalysis is to use a Single-Image Detector (SID) to estimate/quantify if an image is cover or stego, and to average the scores obtained on the set of images.In such a scenario, where Eve does not know the embedding strategies, we experimentally show that if Eve can discriminate among few well-known embedding strategies, she can improve her steganalysis performances compared to a simple averaging or maximum pooled approach. Our discriminative approach allows obtaining steganalysis efficiencies comparable to those obtained by a clairvoyant, Eve, who knows the Alice embedding strategy. Another interesting observation is that DeLS embedding strategy behaves really better than all the other embedding strategies.Those observations results in the experimentation with six different embedding strategies made on Jpeg images with J-UNIWARD, a state-of-the-art Single-Image-Detector, and a discriminative architecture that is invariant to the individual payload in each image, invariant to the size of the analyzed set of images, and build on a binary detector (for the pooling) that is able to deal with various embedding strategies.

show abstract

Steganalysis of Adaptive JPEG Steganography Using 2D Gabor Filters

Cited by 261 publications

References 26 publications

Detection of Classifier Inconsistencies in Image Steganalysis

Detection of Classifier Inconsistencies in Image Steganalysis

CNN-Based Adversarial Embedding for Image Steganography

Pooled Steganalysis in JPEG: how to deal with the spreading strategy?

Contact Info

Product

Resources

About