Nonparametric Steganalysis of QIM Steganography Using Approximate Entropy

The mainstream adaptive steganography algorithm is difficult to resist scaling attack, and the anti-detection performance of the algorithm based on quantization index modulation is not high enough. To solve the above problems, this paper proposes a spatial image adaptive steganography algorithm based on Zernike moment to resist scaling attack and statistic detection. First, the Zernike moment of the normalized cover image is extracted and the new cover is obtained by the dither modulation algorithm. Second, the minimized distortion embedding is realized by Spatial UNIversal WAvelet Relative Distortion (S-UNIWARD) and syndrome-trellis codes coded. And finally, the stego image is generated by changing the original Zernike moment according to the modified amplitude of the new cover. The experimental results show that the proposed algorithm is robust to the three common scaling attacks, and the message extraction error rate is significantly lower than the S-UNIWARD steganography after the scaling attacks; moreover, compared with the steganography algorithm based on quantization index modulation, the proposed algorithm has higher detection resistance.

Robust Video watermarking techniques based on DPCM Encoder and permutated DCT spectrum combined with QIM is introduced. For this purpose the pixels of incoming frames undergo permutation, followed by block wise DCT transform. The DPCM encoder generates differential spectral images. The significant coefficients of these differential spectral images are selected for embedding procedure by quantization index modulation (QIM) method. By this technique the watermark is more robust against attacks and compression. To further increase robustness and to avoid visible degradation additionally the frame order and the watermark pixels were rearranged. To extract the watermark from the video the same operations have to be done on the decoder side. The incoming video frames are permutated followed by block wise DCT transform. After the decoder builds the differential spectral image the embedded coefficients can be selected and watermark extracted by inverse QIM.

Covert channel is a type of secure communications that provides reliable security in transferring information between entities. Quantization Index Modulation (QIM) technique can be used to embed data in images to implement covert communication channels. In single embedding in QIM, each sample is quantized once to carry a single bit of the secret message. Double embedding in QIM was introduced by Sarkar and Manjunath in 2009, to increase the transferred data rate where a single sample is quantized twice to carry two bits in time. In this paper, a higher data transmission rate covert channel is developed using multiple embedding in the QIM framework with multi quantization vectors. This approach considers the embedding process as a communications problem, that it uses a structured scheme of Multiple Trellis-Coded Quantization jointed with Multiple Trellis-Coded Modulation (MTCQ/MTCM) to generate the stego-space. This technique returns a PSNR ≥ 40 dB. More over, increasing the coding states and/or the coding multiplicity returns an altered space with low correlated coefficients, thus improving the transferred data detection error rate performance. To secure the modulation process, the altered coefficients are chaotic scrambled by the 2-D Arnold Cat Map before embedding to increase the uncertainty in case of an attacker discovered the channel.

Quantization index modulation (QIM) principle has shown superior performance in digital watermarking over spread-spectrum and low-bit(s) modulation based methods. In QIM based data hiding scheme, it is a challenging task to embed multiple bits of information into the same set of host signal samples. This work attempts to employ M-ary modulation for multibit QIM watermarking. The watermark embedding process may be divided into two phases. In the first phase, a binary watermark image is spatially dispersed using sequence of number generated by a secret key. In the second phase, host image is decomposed by integer wavelets and the encoded watermark bits are embedded into high-low (HL) and low-high (LH) subbands of DWT-coefficients using M-ary modulation. M-ary modulation is accomplished using near-orthogonal dither generated from the combination of random key, Hadamard matrix and step size. Simulation results show that robustness is increased, at the cost of increase in decoding complexity for high M-value.

Information embedding has recently been recognized as power-limited communication over a "super"-channel with state information non-causally known at the encoder. Based on this equivalence many encoding schemes have been proposed. Quantization index modulation (QIM) family has emerged as an asymptotically information-theoretically optimal embedding function. Optimum transmission rates are achieved when both the watermark-to-noise ratio (WNR) and quantizers dimensionality become sufficiently large. At low WNR however, a trade-off between transmission rate and robustness should be found. This paper is concerned with deriving lattice encoding performances over the resulting modulo lattice additive noise channel (MLAN) for both high and low WNR regimes. To this end, some lattices with good packing and quantizing properties are used and corresponding capacity and bit-error rate (BER) curves are provided for both regular QIM, and distortion-compensated QIM.

Steganography deals with the ways of hiding communicated data in such a way that it remains confidential. Finding best position inside cover image to embed text message, maintaining a reasonable trade-off between security, robustness, higher bit embedding rate and imperceptibility are some of the challenges of steganography system. Hence, this paper presents firefly algorithm for finding best positions inside cover image in order to embed text message into cover image using Pixel Value Differencing (PVD) technique. Four different cover image was used. Experimental result showed the cover image with selected location using firefly algorithm as well as the stego image using PVD technique. The stego image was evaluated using Peak Signal to Noise Ratio (PSNR) and Mean square Error (MSE). Firefly Algorithm with PVD technique produced a promising result for image steganography.

The general purpose of steganalysis algorithms is to detect a secret message in digital images. Nevertheless, previous studies have rather ignored the fact that more valuable information like message length can be extracted from stego images. In this article, a method is proposed for estimating message length in stego images resulted from PQ, which is a common steganography algorithm. In this method, the stego image resulted from PQ algorithm is first calibrated so that the corresponding cover image is obtained. Then, threshold μ is compared to the changes of global histograms of cover and stego images. If the threshold is less than the changes, first-order statistics and global histogram will be used to estimate message length; otherwise, secret message length will be estimated using second-order statistics and changes of variation of cover and stego images.

In this paper, an efficient joint watermarking and Wavelet Scalar Quantization (WSQ) compression scheme based on Quantization Index Modulation (QIM) watermarking approach is proposed to attenuate the impact of the compression operation on the embedded watermark. The proposed scheme allows the embedding of secret message directly into the fingerprint data when compressed by the WSQ algorithm. Experimental results have shown the robustness of the proposed scheme against the WSQ compression while slightly affecting its compression performance (i.e. visual quality and compression rate).

Copyright notification and enforcement, authentication, covert communication, and hybrid transmission applications such as digital audio broadcasting are examples of emerging multimedia applications for digital watermarking and information embedding methods, methods for embedding one signal (e.g., the digital watermark) within another "host" signal to form a third, "composite" signal. The embedding is designed to achieve efficient trade-offs among the three conflicting goals of maximizing information-embedding rate, minimizing distortion between the host signal and composite signal, and maximizing the robustness of the embedding. We present a class of embedding methods called quantization index modulation (QIM) that achieve provably good rate-distortion-robustness performance. These methods, and low-complexity realizations of them called dither modulation, are provably better than both previously proposed linear methods of spread spectrum and nonlinear methods of low-bit(s) modulation against square-error distortion-constrained intentional attacks. We also derive information-embedding capacities for the case of a colored Gaussian host signal and additive colored Gaussian noise attacks. These results imply an information embedding capacity of about 1/3 b/s of embedded digital rate for every Hertz of host signal bandwidth and every dB drop in received host signal quality. We show that QIM methods achieve performance within 1.6 dB of capacity, and we introduce a form of postprocessing we refer to as distortion compensation that, when combined with QIM, allows capacity to be achieved. In addition, we show that distortion-compensated QIM is an optimal embedding strategy against some important classes of intentional attacks as well. Finally, we report simulation results that demonstrate the performance of dither modulation realizations that can be implemented with only a few adders and scalar quantizers.

The objective of steganalysis is to detect messages hidden in a cover images, such as digital images. The ultimate goal of a steganalyst is to extract and decipher the secret message. In this paper, we present a powerful new blind steganalytic scheme that can reliably detect hidden data with a relatively small embedding rate in JPEG images as well as using a technique known as calibration. This would increase the success rate of steganalysis by detecting data in transform domain. This scheme is feature based in the sense that features that are sensitive to embedding changes are being employed as means of steganalysis. The features are extracted in DCT domain. DCT domain features have extended DCT features and Markovian features merged together in calibration technique to eliminate the drawbacks of both(inter and intra block dependency) respectively. For the lesser embedding rate, the features are considered separately to evolve a better classification rate. The blind steganalytic technique has a broad spectrum of analyzing different embedding techniques The feature set contains 274 features by merging both DCT features and Markovian features. The extracted features are being fed to a classifier which helps to distinguish between a cover and stego image. Support Vector Machine is used as classifier here.KeywordsSteganalysisDCTMarkovCalibrationSupport Vector Machine

State-of the-art blind and robust 3D watermarking schemes already withstand combinations of a wide variety of attacks (e.g. Noise addition, simplification, smoothing, etc) except cropping. Spread Spectrum Dithering Modulation (STDM) method is an extension of Quantization Index Modulation (QIM). Besides the simplicity and the trade-off between high capacity and robustness provided by QIM methods, it is also resistant against re-quantization. This paper focuses on two state-of-the-art techniques which offer different and complementary advantages, respectively QIM-based 3D watermarking and feature point-based watermarking synchronization. The idea is to combine both in such a way that the new scheme would benefit from the advantages of both techniques and compensate for their respective fragilities. We show that robustness against cropping and other common attacks is achieved provided that at least one feature point as well as its corresponding local neighborhood is retrieved.

Spread spectrum technique is one way to increase the security and imperceptibility of messages embedded in stego images. The way this method works is by embedding it spreadly and randomly to all images with certain random patterns. Generally, random patterns are implemented using modulation and PN sequence generators. In this research, spread spectrum embedding techniques are implemented with chaotic encryption. Generally, chaotic encryption is done on the message image, but in this research, chaos encryption is applied to the cover image with the aim of spreading the message. Chaos encryption is chosen because this method only randomizes pixel coordinates without changing pixel values. In this way, the message embedding on the smallest bit of the cover image is carried out in a spread with chaotic patterns so that the attacker is not easily guessed. Chaotic patterns will be formed from the Arnold chaotic map formula with parameters determined from password processing, the goal being that patterns are formed more dynamically. In addition, during the embedding process, the message is encrypted with the XOR operation based on the MSB value. Based on testing the proposed method proved to be safer and slightly more imperceptible compared to the previous method.

Image watermarking techniques have continuously evolved to address new challenges and incorporate advanced features. The advent of data-driven approaches has enabled the processing and analysis of large volumes of data, extracting valuable insights and patterns. In this paper, we propose two content-aware quantization index modulation (QIM) algorithms: Content-Aware QIM (CA-QIM) and Content-Aware Minimum Distortion QIM (CAMD-QIM). These algorithms aim to improve the embedding distortion of QIM-based watermarking schemes by considering the statistics of the cover signal vectors and messages. CA-QIM introduces a canonical labeling approach, where the closest coset to each cover vector is determined during the embedding process. An adjacency matrix is constructed to capture the relationships between the cover vectors and messages. CAMD-QIM extends the concept of minimum distortion (MD) principle to content-aware QIM. Instead of quantizing the carriers to lattice points, CAMD-QIM quantizes them to close points in the correct decoding region. Canonical labeling is also employed in CAMD-QIM to enhance its performance. Both schemes can be categorized as (key-aided) semi-blind watermarking. Simulation results demonstrate the effectiveness of CA-QIM and CAMD-QIM in reducing embedding distortion compared to traditional QIM. The combination of canonical labeling and the minimum distortion principle proves to be powerful, minimizing the need for changes to most cover vectors/carriers. These content-aware QIM algorithms provide improved performance and robustness for watermarking applications.

This paper examines the role of attacker's memory in quantization index modulation (QIM) watermarking systems. First we derive the attacker's noise distribution that maximizes probability of error of the detector. Next, we derive QIM code parameters that are minmax optimal. The minmax optimal embedding strategy involves randomized lattice rotations, and the corresponding worst noise distributions are isotropic.

Quantization index modulation (QIM) is widely used in watermarking, but QIM cannot be applied to reversible watermarking directly due to its inherent limitations. In this paper, an improvement is made to QIM to make it suitable for reversible watermarking. Based on the improved QIM (IQIM), a reversible watermarking scheme for 2D CAD engineering graphics is proposed. The relative amplitudes and the relative phases of entity vertices are used as hosts to carry the watermark, respectively, and the data recovery is controlled by a secret key. Experimental results indicate that the proposed scheme can strike a good balance between capacity, imperceptibility, robustness, security and complexity.