Showing papers on "Steganography published in 2018"

HiDDeN: Hiding Data With Deep Networks

Abstract: Recent work has shown that deep neural networks are highly sensitive to tiny perturbations of input images, giving rise to adversarial examples Though this property is usually considered a weakness of learned models, we explore whether it can be beneficial We find that neural networks can learn to use invisible perturbations to encode a rich amount of useful information In fact, one can exploit this capability for the task of data hiding We jointly train encoder and decoder networks, where given an input message and cover image, the encoder produces a visually indistinguishable encoded image, from which the decoder can recover the original message We show that these encodings are competitive with existing data hiding algorithms, and further that they can be made robust to noise: our models learn to reconstruct hidden information in an encoded image despite the presence of Gaussian blurring, pixel-wise dropout, cropping, and JPEG compression Even though JPEG is non-differentiable, we show that a robust model can be trained using differentiable approximations Finally, we demonstrate that adversarial training improves the visual quality of encoded images

Secure Medical Data Transmission Model for IoT-Based Healthcare Systems

Abstract: Due to the significant advancement of the Internet of Things (IoT) in the healthcare sector, the security, and the integrity of the medical data became big challenges for healthcare services applications. This paper proposes a hybrid security model for securing the diagnostic text data in medical images. The proposed model is developed through integrating either 2-D discrete wavelet transform 1 level (2D-DWT-1L) or 2-D discrete wavelet transform 2 level (2D-DWT-2L) steganography technique with a proposed hybrid encryption scheme. The proposed hybrid encryption schema is built using a combination of Advanced Encryption Standard, and Rivest, Shamir, and Adleman algorithms. The proposed model starts by encrypting the secret data; then it hides the result in a cover image using 2D-DWT-1L or 2D-DWT-2L. Both color and gray-scale images are used as cover images to conceal different text sizes. The performance of the proposed system was evaluated based on six statistical parameters; the peak signal-to-noise ratio (PSNR), mean square error (MSE), bit error rate (BER), structural similarity (SSIM), structural content (SC), and correlation. The PSNR values were relatively varied from 50.59 to 57.44 in case of color images and from 50.52 to 56.09 with the gray scale images. The MSE values varied from 0.12 to 0.57 for the color images and from 0.14 to 0.57 for the gray scale images. The BER values were zero for both images, while SSIM, SC, and correlation values were ones for both images. Compared with the state-of-the-art methods, the proposed model proved its ability to hide the confidential patient’s data into a transmitted cover image with high imperceptibility, capacity, and minimal deterioration in the received stego-image.

Deep residual learning for image steganalysis

Abstract: Image steganalysis is to discriminate innocent images and those suspected images with hidden messages. This task is very challenging for modern adaptive steganography, since modifications due to message hiding are extremely small. Recent studies show that Convolutional Neural Networks (CNN) have demonstrated superior performances than traditional steganalytic methods. Following this idea, we propose a novel CNN model for image steganalysis based on residual learning. The proposed Deep Residual learning based Network (DRN) shows two attractive properties than existing CNN based methods. First, the model usually contains a large number of network layers, which proves to be effective to capture the complex statistics of digital images. Second, the residual learning in DRN preserves the stego signal coming from secret messages, which is extremely beneficial for the discrimination of cover images and stego images. Comprehensive experiments on standard dataset show that the DRN model can detect the state of arts steganographic algorithms at a high accuracy. It also outperforms the classical rich model method and several recently proposed CNN based methods.

Image steganography in spatial domain: A survey

Abstract: This paper presents a literature review of image steganography techniques in the spatial domain for last 5 years. The research community has already done lots of noteworthy research in image steganography. Even though it is interesting to highlight that the existing embedding techniques may not be perfect, the objective of this paper is to provide a comprehensive survey and to highlight the pros and cons of existing up-to-date techniques for researchers that are involved in the designing of image steganographic system. In this article, the general structure of the steganographic system and classifications of image steganographic techniques with its properties in spatial domain are exploited. Furthermore, different performance matrices and steganalysis detection attacks are also discussed. The paper concludes with recommendations and good practices drawn from the reviewed techniques.

A Novel Image Steganography Method via Deep Convolutional Generative Adversarial Networks

Abstract: The security of image steganography is an important basis for evaluating steganography algorithms. Steganography has recently made great progress in the long-term confrontation with steganalysis. To improve the security of image steganography, steganography must have the ability to resist detection by steganalysis algorithms. Traditional embedding-based steganography embeds the secret information into the content of an image, which unavoidably leaves a trace of the modification that can be detected by increasingly advanced machine-learning-based steganalysis algorithms. The concept of steganography without embedding (SWE), which does not need to modify the data of the carrier image, appeared to overcome the detection of machine-learning-based steganalysis algorithms. In this paper, we propose a novel image SWE method based on deep convolutional generative adversarial networks. We map the secret information into a noise vector and use the trained generator neural network model to generate the carrier image based on the noise vector. No modification or embedding operations are required during the process of image generation, and the information contained in the image can be extracted successfully by another neural network, called the extractor, after training. The experimental results show that this method has the advantages of highly accurate information extraction and a strong ability to resist detection by state-of-the-art image steganalysis algorithms.

A fusion steganographic algorithm based on Faster R-CNN

Abstract: The aim of information hiding is to embed the secret information in a normal cover media such as image, video, voice or text, and then transmit the secret data through the transmission of the public information. The secret message should not be damaged when the processing is applied on the cover media. In order to ensure the invisibility of confidential information, complex texture objects should be suitable for embedding information. In this paper, an approach which corresponds multiple steganographic algorithms to complex texture objects was presented for hiding secret data. Firstly, complex texture regions based on objects detection are selected. Secondly, several different steganographic methods were used to hide secret data into the selected area block. Experimental results show that this approach enhances the security and provides robust embedding.

A new 4-D chaotic hyperjerk system, its synchronization, circuit design and applications in RNG, image encryption and chaos-based steganography

Abstract: Hyperjerk systems have received significant interest in the literature because of their simple structure and complex dynamical properties This work presents a new chaotic hyperjerk system having two exponential nonlinearities Dynamical properties of the chaotic hyperjerk system are discovered through equilibrium point analysis, bifurcation diagram, dissipativity and Lyapunov exponents Moreover, an adaptive backstepping controller is designed for the synchronization of the chaotic hyperjerk system Also, a real circuit of the chaotic hyperjerk system has been carried out to show the feasibility of the theoretical hyperjerk model The chaotic hyperjerk system can also be useful in scientific fields such as Random Number Generators (RNGs), data security, data hiding, etc In this work, three implementations of the chaotic hyperjerk system, viz RNG, image encryption and sound steganography have been performed by using complex dynamics characteristics of the system

An overview of steganography techniques applied to the protection of biometric data

Abstract: Identification of persons by way of biometric features is an emerging phenomenon. Over the years, biometric recognition has received much attention due to its need for security. Amongst the many existing biometrics, fingerprints are considered to be one of the most practical ones. Techniques such as watermarking and steganography have been used in attempt to improve security of biometric data. Watermarking is the process of embedding information into a carrier file for the protection of ownership/copyright of music, video or image files, whilst steganography is the art of hiding information. This paper presents an overview of steganography techniques applied in the protection of biometric data in fingerprints. It is novel in that we also discuss the strengths and weaknesses of targeted and blind steganalysis strategies for breaking steganography techniques.

Invisible Steganography via Generative Adversarial Networks

Abstract: Nowadays, there are plenty of works introducing convolutional neural networks (CNNs) to the steganalysis and exceeding conventional steganalysis algorithms. These works have shown the improving potential of deep learning in information hiding domain. There are also several works based on deep learning to do image steganography, but these works still have problems in capacity, invisibility and security. In this paper, we propose a novel CNN architecture named as \isgan to conceal a secret gray image into a color cover image on the sender side and exactly extract the secret image out on the receiver side. There are three contributions in our work: (i) we improve the invisibility by hiding the secret image only in the Y channel of the cover image; (ii) We introduce the generative adversarial networks to strengthen the security by minimizing the divergence between the empirical probability distributions of stego images and natural images. (iii) In order to associate with the human visual system better, we construct a mixed loss function which is more appropriate for steganography to generate more realistic stego images and reveal out more better secret images. Experiment results show that ISGAN can achieve start-of-art performances on LFW, Pascal VOC2012 and ImageNet datasets.

A review of image steganalysis techniques for digital forensics

Abstract: Steganalysis and steganography are the two different sides of the same coin. Steganography tries to hide messages in plain sight while steganalysis tries to detect their existence or even more to retrieve the embedded data. Both steganography and steganalysis received a great deal of attention, especially from law enforcement. While cryptography in many countries is being outlawed or limited, cyber criminals or even terrorists are extensively using steganography to avoid being arrested with encrypted incriminating material in their possession. Therefore, understanding the ways that messages can be embedded in a digital medium –in most cases in digital images-, and knowledge of state of the art methods to detect hidden information, is essential in exposing criminal activity. Digital image steganography is growing in use and application. Many powerful and robust methods of steganography and steganalysis have been presented in the literature over the last few years. In this literature review, we will discuss and present various steganalysis techniques – from earlier ones to state of the art- used for detection of hidden data embedded in digital images using various steganography techniques.

Adversarial Examples Against Deep Neural Network based Steganalysis

Abstract: Deep neural network based steganalysis has developed rapidly in recent years, which poses a challenge to the security of steganography. However, there is no steganography method that can effectively resist the neural networks for steganalysis at present. In this paper, we propose a new strategy that constructs enhanced covers against neural networks with the technique of adversarial examples. The enhanced covers and their corresponding stegos are most likely to be judged as covers by the networks. Besides, we use both deep neural network based steganalysis and high-dimensional feature classifiers to evaluate the performance of steganography and propose a new comprehensive security criterion. We also make a tradeoff between the two analysis systems and improve the comprehensive security. The effectiveness of the proposed scheme is verified with the evidence obtained from the experiments on the BOSSbase using the steganography algorithm of WOW and popular steganalyzers with rich models and three state-of-the-art neural networks.

Image steganography based on Canny edge detection, dilation operator and hybrid coding

Abstract: In this paper, a novel steganography algorithm based on local reference edge detection technique and exclusive disjunction (XOR) property is proposed. Human eyes are less sensitive towards intensity changes in the sharp edge region compared to the uniform region of the image. Because of this, the secret message bits have been embedded in the sharp regions by local reference pixels which are detected by Canny edge method and optimized by dilation morphological operator. The predefined sets of pixels are easily identified with less computational complexity in the stego image. The embedding algorithm improved in terms of security and capacity using bit plane dependent XOR coding technique that makes least possible alterations in LSB bits of edge pixels. The existing edge-based steganography techniques provide better imperceptibility but relatively limits the embedding capacity. The proposed method efficiently improves the embedding capacity with an acceptable range of imperceptibility and robustness. The simulation results evaluated using full reference image quality assessment method, it exhibits better embedding capacity (bpp) compared to existing steganography techniques retaining the values of PSNR and structural similarity (SSIM).

RNN-SM: Fast Steganalysis of VoIP Streams Using Recurrent Neural Network

Abstract: Quantization index modulation (QIM) steganography makes it possible to hide secret information in voice-over IP (VoIP) streams, which could be utilized by unauthorized entities to set up covert channels for malicious purposes. Detecting short QIM steganography samples, as is required by real circumstances, remains an unsolved challenge. In this paper, we propose an effective online steganalysis method to detect QIM steganography. We find four strong codeword correlation patterns in VoIP streams, which will be distorted after embedding with hidden data. To extract those correlation features, we propose the codeword correlation model, which is based on recurrent neural network (RNN). Furthermore, we propose the feature classification model to classify those correlation features into cover speech and stego speech categories. The whole RNN-based steganalysis model (RNN-SM) is trained in a supervised learning framework. Experiments show that on full embedding rate samples, RNN-SM is of high detection accuracy, which remains over 90% even when the sample is as short as 0.1 s, and is significantly higher than other state-of-the-art methods. For the challenging task of conducting steganalysis towards low embedding rate samples, RNN-SM also achieves a high accuracy. The average testing time for each sample is below 0.15% of sample length. These clues show that RNN-SM meets the short sample detection demand and is a state-of-the-art algorithm for online VoIP steganalysis.

Information hiding in edges: A high capacity information hiding technique using hybrid edge detection

Abstract: The multimedia security is becoming more and more important as the data being exchanged on the Internet is increasing exponentially. Though cryptography is one of the methods which is used to secure the data during transit, but the camouflaged appearance of the scrambled data alerts the adversary about some critical information being shared. In such a scenario, steganography has been used as an alternate solution to secure the secret information. In this paper a color image steganographic algorithm based on hybrid edge detection is proposed. The color image is partitioned into constituent Red (R), Green (G) and Blue (B) planes. Hybrid edge detection is used for finding the edge and non-edge pixels of Green and Blue planes of cover image. The Green and Blue planes are used for hiding the data while Red plane holds the pixel status (whether edge or non-edge) of these planes. The RC4 encryption algorithm is used to encrypt secret message before embedding it in the cover image to enhance security of the secret data. A fragile watermark/logo (whose size is less than 1% of total secret data) has been embedded, besides secret data in the cover image, to facilitate content authentication and early tamper detection. At the receiver, firstly logo is extracted. If it is same as one embedded at transmitter, indicating that secret data has not been altered during transit, secret data is extracted. Otherwise (if extracted logo is not same as used at input) the receiver does not waste critical time to extract compromised data but sends an automatic retransmission request. Experimental investigations reveal that the proposed scheme is capable of providing high quality of stego-images for a fairly high pay load. A comparison of the proposed technique with some state of art schemes substantiates the above arguments.

A modified DWT-based image steganography technique

Abstract: The main challenges of image steganography are imperceptibility of the cover image and no recoverability of the secret data. To deal with these challenges, a modified digital image steganography technique based on Discrete Wavelet Transform (DWT) is proposed. In proposed approach, two new concepts are being proposed to minimize the distortion in the cover image. The first one i.e. secret key computation concept is proposed to make it more robust and resistive towards steganalysis. The second one, known as blocking concept, is introduced to ensure least variation in the cover image. The proposed approach is tested over ten different cover images and two secret images. Its performance is compared with the six well-known steganography techniques. The experimental results reveal that the proposed approach performs better than the existing techniques in terms of imperceptibility, security and quality measures. The six image processing attacks are also applied on the stego-image to test the robustness of the proposed approach. The effects of compression, rotation, and application of different wavelets have also been investigated on the proposed approach. The results demonstrate the robustness of the proposed approach under different image processing attacks. Both stego-image and extracted secret images possess better visual quality.

StegNet: Mega Image Steganography Capacity with Deep Convolutional Network.

Abstract: Traditional image steganography often leans interests towards safely embedding hidden information into cover images with payload capacity almost neglected. This paper combines recent deep convolutional neural network methods with image-into-image steganography. It successfully hides the same size images with a decoding rate of 98.2% or bpp (bits per pixel) of 23.57 by changing only 0.76% of the cover image on average. Our method directly learns end-to-end mappings between the cover image and the embedded image and between the hidden image and the decoded image. We~further show that our embedded image, while with mega payload capacity, is still robust to statistical analysis.

High capacity, transparent and secure audio steganography model based on fractal coding and chaotic map in temporal domain

Abstract: Information hiding researchers have been exploring techniques to improve the security of transmitting sensitive data through an unsecured channel. This paper proposes an audio steganography model for secure audio transmission during communication based on fractal coding and a chaotic least significant bit or also known as HASFC. This model contributes to enhancing the hiding capacity and preserving the statistical transparency and security. The HASFC model manages to embed secret audio into a cover audio with the same size. In order to achieve this result, fractal coding is adopted which produces high compression ratio with the acceptable reconstructed signal. The chaotic map is used to randomly select the cover samples for embedding and its initial parameters are utilized as a secret key to enhancing the security of the proposed model. Unlike the existing audio steganography schemes, The HASFC model outperforms related studies by improving the hiding capacity up to 30% and maintaining the transparency of stego audio with average values of SNR at 70.4, PRD at 0.0002 and SDG at 4.7. Moreover, the model also shows resistance against brute-force attack and statistical analysis.

Enhancing PC Data Security via Combining RSA Cryptography and Video Based Steganography

Abstract: This paper proposed an enhanced system for securing sensitive text-data on personal computer benefitting from the combination of both techniques: cryptography and steganography. The system security is generated by involving RSA cryptography followed by video based steganography as two sequential layers to insure best possible security gaining the advantages from both. The study modeled the system and implemented it to be tested to explore the relation between security, capacity and data dependency. The experimentations covered testing securing data within 15 different size videos showing interesting results. The research gave enhancement to capacity vs. security, as enforced unavoidable tradeoff. The work uniqueness is presented in showing different measures allowing the user and application to be the decision maker to choose. The tests provided all possibilities of accepting security of 1-LSB, 2-LSB, and 3-LSB methods detailing their effects on the cover video. The main outcome proved applicability to be adopting 3-LSB approach to give acceptable security with practical capacity preferred making 3-LSB winning among 1-LSB and 2-LSB techniques.

New LSB-based colour image steganography method to enhance the efficiency in payload capacity, security and integrity check

Abstract: Steganography is the technique for hiding information within a carrier file so that it is imperceptible for unauthorized parties. In this study, it is intended to combine many techniques to gather a new method for colour image steganography to obtain enhanced efficiency, attain increased payload capacity, posses integrity check and security with cryptography at the same time. Proposed work supports many different formats as payload. In the proposed method, the codeword is firstly formed with secret data and its CRC-32 checksum, then the codeword is compressed by Gzip just before encrypting it by AES, and it is finally added to encrypted header information for further process and then embedded into the cover image. Embedding the encrypted data and header information process utilizes Fisher-Yates Shuffle algorithm for selecting next pixel location. To hide one byte, different LSB (least significant bits) of all colour channels of the selected pixel is exploited. In order to evaluate the proposed method, comparative performance tests are carried out against different spatial image steganographic techniques using some of the well-known image quality metrics. For security analysis, histogram, enhanced LSB and Chi-square analyses are carried out. The results indicate that with the proposed method has an improved payload capacity, security and integrity check for common problems of simple LSB method. Moreover, it has been shown that the proposed method increases the visual quality of the stego image when compared to other studied methods, and makes the secret message difficult to be discovered.

Image steganalysis using improved particle swarm optimization based feature selection

Abstract: Image steganalysis is the task of discovering the hidden message in a multimedia file in which the steganalysis technique highly depends on the feature elements of the image. Since there is a possibility for a feature vector to contain redundant elements, processing of redundant elements can be harmful in terms of long computation cost and large storage space. This paper proposes a new feature selection approach based on Adaptive inertia weight-based Particle Swarm Optimization (APSO) for the image steganalysis where the inertia weight of PSO is adaptively adjusted using the swarm diameter, average distance of particles around the center and average speed of particles towards the center. Also, the proposed APSO is used with the novel measure of Area Under the receiver operating characteristics Curve (AUC) as the fitness function to enhance the performance of identification of stego-images from the cover images in steganalysis problem. Due to appropriate convergence rate and the regulated search step of APSO, it is able to select the most significant and influential feature elements and so, the performance of steganalysis will be improved. Experimental results of the proposed method on Breaking Out Steganography System (BOSS) benchmark proves the superiority of the proposed method compared to the similar approaches in image steganalysis in terms of detection of stego-image, running time, and diversity measure.

CNN Based Adversarial Embedding with Minimum Alteration for Image Steganography.

Abstract: 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.

A Comparative Analysis of Information Hiding Techniques for Copyright Protection of Text Documents

Abstract: With the ceaseless usage of web and other online services, it has turned out that copying, sharing, and transmitting digital media over the Internet are amazingly simple. Since the text is one of the main available data sources and most widely used digital media on the Internet, the significant part of websites, books, articles, daily papers, and so on is just the plain text. Therefore, copyrights protection of plain texts is still a remaining issue that must be improved in order to provide proof of ownership and obtain the desired accuracy. During the last decade, digital watermarking and steganography techniques have been used as alternatives to prevent tampering, distortion, and media forgery and also to protect both copyright and authentication. This paper presents a comparative analysis of information hiding techniques, especially on those ones which are focused on modifying the structure and content of digital texts. Herein, various text watermarking and text steganography techniques characteristics are highlighted along with their applications. In addition, various types of attacks are described and their effects are analyzed in order to highlight the advantages and weaknesses of current techniques. Finally, some guidelines and directions are suggested for future works.

Spatial Image Steganography Based on Generative Adversarial Network.

Abstract: With the recent development of deep learning on steganalysis, embedding secret information into digital images faces great challenges. In this paper, a secure steganography algorithm by using adversarial training is proposed. The architecture contain three component modules: a generator, an embedding simulator and a discriminator. A generator based on U-NET to translate a cover image into an embedding change probability is proposed. To fit the optimal embedding simulator and propagate the gradient, a function called Tanh-simulator is proposed. As for the discriminator, the selection-channel awareness (SCA) is incorporated to resist the SCA based steganalytic methods. Experimental results have shown that the proposed framework can increase the security performance dramatically over the recently reported method ASDL-GAN, while the training time is only 30% of that used by ASDL-GAN. Furthermore, it also performs better than the hand-crafted steganographic algorithm S-UNIWARD.

Multi-Bits Stego-System For Hiding Text in Multimedia Images Based on User Security Priority

Abstract: Security system for hiding sensitive text-data on personal computer (PC) is implemented. The system proposes using normal multimedia image based steganography replacing the pixel least significant bits with text to be hidden. The study involved several testing scenarios for possible increase in the capacity and ambiguity within steganography adopting 1, 2, 3 and 4 least significant bits stego-systems choices. The design novelty is in providing full security information to the user to select appropriate cover-image from the PC based on the security priority. The technique allows the PC user to test multi-bits steganography on several images to hide same sensitive texts. Then, the user can prefer one image to be used as cover image based on his selection knowing the capacity and security priority desired. The study proofs the data dependency and its security property by experimenting 35 fixed sizes PC images showing remarkable results.

End-to-End Trained CNN Encoder-Decoder Networks for Image Steganography

Abstract: All the existing image steganography methods use manually crafted features to hide binary payloads into cover images. This leads to small payload capacity and image distortion. Here we propose a convolutional neural network based encoder-decoder architecture for embedding of images as payload. To this end, we make following three major contributions: (i) we propose a deep learning based generic encoder-decoder architecture for image steganography; (ii) we introduce a new loss function that ensures joint end-to-end training of encoder-decoder networks; (iii) we perform extensive empirical evaluation of proposed architecture on a range of challenging publicly available datasets (MNIST, CIFAR10, PASCAL-VOC12, ImageNet, LFW) and report state-of-the-art payload capacity at high PSNR and SSIM values.