Steganalysis Scheme Research Articles

Steganalysis of content-adaptive binary image data hiding

Most state-of-the-art binary image data hiding methods concentrate the embedding changes on the centers of l-shape patterns. This embedding criterion, however, introduces an unbalanced modification on boundary structures. This paper proposes a steganalytic scheme to detect recently developed content-adaptive binary image data hiding by exploiting the embedding effect associated with the l-shape pattern-based embedding criterion. We first assess how changing l-shape patterns affects the distribution of a special 4×3 sized pattern. Based on the assessment, 4 classes of patterns that model the distribution of two pixels oriented the direction of pattern changing are employed to define a 32-dimensional steganalytic feature set. Experimental results show that, despite of the low dimensionality, the proposed steganalytic features can effectively detect state-of-the-art binary image data hiding schemes, especially those pattern-tracing-based approaches.

Detecting Steganography of Adaptive Multirate Speech with Unknown Embedding Rate

Steganalysis of adaptive multirate (AMR) speech is a significant research topic for preventing cybercrimes based on steganography in mobile speech services. Differing from the state-of-the-art works, this paper focuses on steganalysis of AMR speech with unknown embedding rate, where we present three schemes based on support-vector-machine to address the concern. The first two schemes evolve from the existing image steganalysis schemes, which adopt different global classifiers. One is trained on a comprehensive speech sample set including original samples and steganographic samples with various embedding rates, while the other is trained on a particular speech sample set containing original samples and steganographic samples with uniform distributions of embedded information. Further, we present a hybrid steganalysis scheme, which employs Dempster–Shafer theory (DST) to fuse all the evidence from multiple specific classifiers and provide a synthesized detection result. All the steganalysis schemes are evaluated using the well-selected feature set based on statistical characteristics of pulse pairs and compared with the optimal steganalysis that adopts specialized classifiers for corresponding embedding rates. The experimental results demonstrate that all the three steganalysis schemes are feasible and effective for detecting the existing steganographic methods with unknown embedding rates in AMR speech streams, while the DST-based scheme outperforms the others overall.

Segmentation Based Video Steganalysis to Detect Motion Vector Modification

This paper presents a steganalytic approach against video steganography which modifies motion vector (MV) in content adaptive manner. Current video steganalytic schemes extract features from fixed-length frames of the whole video and do not take advantage of the content diversity. Consequently, the effectiveness of the steganalytic feature is influenced by video content and the problem of cover source mismatch also affects the steganalytic performance. The goal of this paper is to propose a steganalytic method which can suppress the differences of statistical characteristics caused by video content. The given video is segmented to subsequences according to block’s motion in every frame. The steganalytic features extracted from each category of subsequences with close motion intensity are used to build one classifier. The final steganalytic result can be obtained by fusing the results of weighted classifiers. The experimental results have demonstrated that our method can effectively improve the performance of video steganalysis, especially for videos of low bitrate and low embedding ratio.

Spatial Steganalysis Using Contrast of Residuals

This letter proposes a novel scheme for spatial steganalysis based on contrast of residuals (CoR). After selecting complex blocks from an uncompressed image by a fluctuation function, the residuals are calculated from the selected blocks and the whole image after applying diverse filters. The CoR is represented as an angle and the norm of residuals is considered as the corresponding weight of angle, which is used as the new steganalysis feature. In the proposed scheme, no quantization and truncation is required and the effective information of long-range dependencies among pixels is kept properly. Also, the dimensionality of feature is linear with the number of residuals. The accuracy of proposed scheme is evaluated on HUGO and WOW algorithms, and the experimental results show that the proposed CoR feature has superior performance at low embedding rate with lower dimensionality.

Performance improvement of LSB-based steganalysis using bit-plane decomposition of images

In this paper, we present an improved least significant bit (LSB)-based steganalysis scheme using the bit-plane decomposition of images. We derive a mathematical condition that can enhance the detection rate for hidden messages based on the correlation coefficient between two parts of a decomposed image. Based on this condition, images are classified and segregated into two groups: the full image including all of the bit-planes and a sub-image containing only the lower bit-planes. The feature vectors for steganalysis are extracted independently form each group. Three types of conventional feature vectors were extracted to verify our proposed method and experiments demonstrated that conventional steganalysis schemes exhibited improved performance using our proposed method. In conclusion, our scheme can be used as a general steganalyzer regardless of the specific steganalysis methods employed for LSB-based steganalysis.

Adaptive Steganalysis Based on Embedding Probabilities of Pixels

In modern steganography, embedding modifications are highly concentrated on the textural regions within an image, as such regions are difficult to model for steganalysis. Previous studies have shown that compared with non-adaptive strategies, this content adaptive strategy achieves stronger security against existing steganalysis. Based on the experiments and analyses, however, we found that this embedding property would inevitably lead to a large limitation in existing adaptive steganography. That is, it is possible for steganalyzers to estimate the regions that have probably been modified after data hiding. In this paper, we propose an adaptive steganalytic scheme based on embedding probabilities of pixels. The main idea of our scheme is that we assign different weights to different pixels in feature extraction. For those pixels with high embedding probabilities, their corresponding weights are larger, since they should contribute more to steganalysis and vice versa. By doing so, we can concentrate our attention on the regions that have probably been modified and significantly reduce the impact of other unchanged smooth regions. It is expected that our proposed method is an improvement on the existing steganalytic methods, which usually assume every pixel has the same contribution to steganalysis. The extensive experiments evaluated on four typical adaptive steganographic methods have shown the effectiveness of the proposed scheme, especially for low embedding rates, for example, lower than 0.20 bpp.

Digital Video Steganalysis Based on a Spatial Temporal Detector

This paper presents a novel digital video steganalysis scheme against the spatial domain video steganography technology based on a spatial temporal detector (ST_D) that considers both spatial and temporal redundancies of the video sequences simultaneously. Three descriptors are constructed on XY, XT and YT planes respectively to depict the spatial and temporal relationship between the current pixel and its adjacent pixels. Considering the impact of local motion intensity and texture complexity on the histogram distribution of three descriptors, each frame is segmented into non-overlapped blocks that are 8×8 in size for motion and texture analysis. Subsequently, texture and motion factors are introduced to provide reasonable weights for histograms of the three descriptors of each block. After further weighted modulation, the statistics of the histograms of the three descriptors are concatenated into a single value to build the global description of ST_D. The experimental results demonstrate the great advantage of our features relative to those of the rich model (RM), the subtractive pixel adjacency model (SPAM) and subtractive prediction error adjacency matrix (SPEAM), especially for compressed videos, which constitute most Internet videos.

Audio steganalysis based on reversed psychoacoustic model of human hearing

During the last decade, audio information hiding has attracted lots of attention due to its ability to provide a covert communication channel. On the other hand, various audio steganalysis schemes have been developed to detect the presence of any secret messages. Basically, audio steganography methods attempt to hide their messages in areas of time or frequency domains where human auditory system (HAS) does not perceive. Considering this fact, we propose a reliable audio steganalysis system based on the reversed Mel-frequency cepstral coefficients (R-MFCC) which aims to provide a model with maximum deviation from HAS model. Genetic algorithm is deployed to optimize dimension of the R-MFCC-based features. This will both speed up feature extraction and reduce the complexity of classification. The final decision is made by a trained support vector machine (SVM) to detect suspicious audio files. The proposed method achieves detection rates of 97.8% and 94.4% in the targeted (Steghide@1.563%) and universal scenarios. These results are respectively 17.3% and 20.8% higher than previous D2-MFCC based method.

Digital video steganalysis toward spread spectrum data hiding

In this study, the authors propose a steganalytic scheme for digital video spread spectrum (SS) data hiding. The proposed method estimates both the hidden message and the gain factor of the SS embedding rules. In this method, the cover frames are first estimated and are compared with the received video frames. Then, the residual matrix is computed and specific features are extracted from this matrix as well as the video frames and estimated frames. The support vector machine, then applies to the extracted features to classify the video as either clean or suspicious. If the video is declared suspicious, both the hidden message and the embedding process gain factor are estimated and consequently the original video is reconstructed. The simulation results confirm the success of the authors’ proposed method in detecting the stego video, estimation of the hidden message and gain factor as well as reconstruction of the original video.

Distributed steganalysis of compressed speech

In this paper, we present a distributed steganalysis scheme for compressed speech in voice-over-IP scenarios to provide fast and precise detection results. In this scheme, each speech parameter available for concealing information is designed to be detected independently exploiting the corresponding optimal detection feature. To achieve this purpose, we introduce four detection features, including histogram distribution, differential histogram distribution, Markov transition matrix and differential Markov transition matrix. These features stem from both long-time distribution characteristics and short-time invariance characteristics of speech signals. We evaluate their performance for steganalysis based on support vector machines with a large number of steganographic G.729a speech samples at different embedding rates or with various sample lengths and compare them with some existing algorithms. The experimental results demonstrate that the presented algorithms can offer excellent steganalysis performance for all speech parameters in any case and outperform the previous ones. Moreover, it is proved that the four features have diverse performance for steganalysis of different speech parameters, which suggests that it is feasible to achieve the distributed steganalysis employing the optimal feature to detect the corresponding parameter in a faster and more efficient manner.

Spatial steganalysis using redistributed residuals and diverse ensemble classifier

This paper proposes a spatial steganalytic scheme based on redistributed residuals and diverse ensemble classifier. In the scheme, the residuals obtained by local pixel predictors are shifted and suppressed to form the redistributed residuals, and then 2929 dimension features are calculated from the first-order statistic of original and redistributed residuals. With the aid of first-order statistic in a broad range, the feature can preserve the long-range dependencies between pixels, especially for highly adaptive steganography. Another advantage of the feature is that the dimension of first-order statistic is linear with range of residuals, so that the computational complexity is lowered. Moreover, with Bagging and AdaBoost mechanisms introduced here, we can get diverse representation of final feature and enhance the individual weak classifier, respectively. Compared with previous works, experimental results show that our scheme is effective at low embedding rate and is characterized by lower computational complexity.

Towards blind detection of low‐rate spatial embedding in image steganalysis

Steganalysis of least significant bit (LSB) embedded images in spatial domain has been investigated extensively over the past decade and most well-known LSB steganography methods have been shown to be detectable. However, according to the latest findings in the area, two major issues of very low-rate (VLR) embedding and content-adaptive steganography have remained hard to resolve. The problem of VLR embedding is indeed a generic problem to any steganalyser, while the issue of adaptive embedding specifically depends on the hiding algorithm employed. The latter challenge has recently been brought up again to the area of LSB steganalysis by highly undetectable stego image steganography that offers a content-adaptive embedding scheme for grey-scale images. The authors new image steganalysis method suggests analysis of the relative norm of the image Clouds manipulated in an LSB embedding system. The method is a self-dependent image analysis and is capable of operating on low-resolution images. The proposed algorithm is applied to the image in spatial domain through image Clouding, relative auto-decorrelation features extraction and quadratic rate estimation, as the main steps of the proposed analysis procedure. The authors then introduce and use new statistical features, Clouds-Min-Sum and Local-Entropies-Sum, which improve both the detection accuracy and the embedding rate estimation. They analytically verify the functionality of the scheme. Their simulation results show that the proposed approach outperforms some well known, powerful LSB steganalysis schemes, in terms of true and false detection rates and mean squared error.

Digital image steganalysis based on local textural features and double dimensionality reduction

This work proposes a spatial steganalysis scheme based on local textural features and double dimensionality reduction. First, an image is filtered by multiple filters to obtain a number of residual images. Local textural patterns are obtained by comparing the pixel values with the neighbors' value in each residual image. By combining all local textural patterns, a high-dimensional textural feature set is formed. Then, principal component analysis is used to perform double dimensionality reduction for high-dimensional textural features. In the first dimensionality reduction stage, the correlation from the same filter is eliminated, while the correlation from different filters can be also eliminated in the second dimensionality reduction stage. Finally, a textural feature set with low dimensionality is proposed and can be effectively used in steganalysis. Experimental results show that proposed textural feature set can efficiently detect adaptive steganographic schemes in spatial domain. Copyright © 2014 John Wiley & Sons, Ltd.

The Analysis of Steganographic by Sub-Pixel Calibration

In order to achieve high accuracy, we present a new calibration technique aimed at blind steganalysis. The calibration can be considered as a preprocessing of stego image before extracting the features. The extracted features will be more effective for our classification [1] Moreover, the calibrated feature was used to train SVM (Support Vector Machine)[2], a nonlinear classifier, which is effective in class separation. For comparison, we conducted extensive experiments and drawn a conclusion that the steganalytic scheme based on our novel calibration can detect the stego information with high accuracy. Experimental results demonstrate that our proposed scheme outperforms the best effective JPEG steganalysis having been presented.

Location-Based Image Steganography

The objective of steganography is to conceal the presence of a secret communication. Nevertheless, with the development of steganalytic attack, also known as steganalysis, there are many statistical methods to estimate the presence and ratio of secret messages. The chi square (χ2) attack and regular-singular (RS) attack are two well recognized and widely applied statistical steganalysis schemes. In this paper, a new image steganography scheme with location selection is proposed to resist statistical steganalysis. The experiment results indicate that the proposed method can resist both χ2 attacks as well as RS attacks.

Steganalysis based on distribution characters of stego-images in reduced dimension space

In this paper we propose an Improved Kernel Linear Discriminant Analysis algorithm to analyze the distribution differences between cover images and stego-images in the reduced dimensional space. We observe that the hidden information, the information hidden in the cover images, of stego-images are clustered in a plane while all other information of cover images are scattered more evenly in the whole space and have no other clusters. Based on this fact, we develop a steganalysis scheme to discriminate stego-images from innocent images. The experiment results show the effectiveness of the propose approach.

A New Steganographic Method For Grayscale Image Using Graph Coloring Problem

Statistical steganalysis schemes detect the existence of secret information embedded by steganography. The 2 detection and Regular-Singular (RS)-attack methods are two well known statistical steganalysis schemes used against LSB (least significant bit) steganography. The embedded message length can be estimated accurately by these two steganalysis schemes. For secret communi- cation, the resistance of steganography against steganalysis is very important for information security. To avoid the enemy's attempts, the statistical features between stego-images and cover images should be as similar as possible for better resistance to steganalysis. To ensure the security against the RS and 2 analysis, we presents in this paper a new steganographic method based on graph coloring problem (GCP). Before embedding the secret message in LSB (least significant bit) of the cover image, we use a (GCP) algorithm to locate the optimal positions of the pixels in the cover image. Thus, the existence of the secret message is hard to be detected by the RS analysis. Meanwhile, better visual quality can be achieved by the proposed algorithm. The experimental results demonstrate the proposed algorithm's effectiveness in resistance to steganalysis with better visual quality.

JPEG Steganalysis With High-Dimensional Features and Bayesian Ensemble Classifier

This work proposes a JPEG steganalytic scheme based on high-dimensional features and Bayesian ensemble classifier. The proposed scheme employs 15700 dimension features calculated from the co-occurrence matrices of DCT coefficients and coefficient differences, which indicate the intra-block and inter-block dependencies of image content. Furthermore, a number of sub-classifiers trained on the features are integrated as an ensemble classifier with a Bayesian mechanism, which is used to give optimal decisions for suspicious images. Experimental results show that both the high-dimensional features and the Bayesian mechanism contribute to the extended scheme, and the performance of the extended scheme is better than those of previous schemes.

Active steganalysis for interpolation-error based reversible data hiding

The interpolation-error based (IEB) reversible data hiding combines difference expansion and histogram-shifting techniques to embed secret data into interpolation-error histograms of a stego-image with high payload and low distortion has been proposed recently. In this paper, an active steganalysis scheme is proposed by analyzing and modeling histogram abnormality in the interpolation-error domain of sub-sampled images with the general Gaussian distribution (GGD) features. A support vector machine (SVM) classifier is trained by estimated parameters of GGD, and then a quantitative algorithm is presented to estimate the embedding length and locations. The experimental results show that the proposed active steganalysis scheme is effective in not only detecting the IEB method but also estimating its message length and embedding locations.

JPEG Steganalysis Based on Feature Fusion by Principal Component Analysis

Aiming to the problems in the existing JPEG steganalysis schemes, such as high redundancy in features and failure to make good use of the complementarities among them, this study proposed a JPEG steganalysis approach based on feature fusion by the principal component analysis (PCA) and analysis of the complementarities among features. The study fused complementary features and isolated redundant components by PCA, and finally used RBaggSVM classifier for classification. Experimental results show that this scheme effectively improves the detection rate of steganalysis in JPEG images and achieves faster speed of image classification.