Xinpeng Zhang

Vulnerability of pixel-value differencing steganography to histogram analysis and modification for enhanced security

The pixel-value differencing (PVD) steganography can embed a large amount of secret bits into a still image with high imperceptibility as it makes use of the characteristics of human vision sensitivity. However, a loophole exists in the PVD method. Unusual steps in the histogram of pixel differences reveal the presence of a secret message. An analyst can even estimate the length of hidden bits from the histogram. To enhance security, a modified scheme is proposed which avoids occurrence of the above-mentioned steps in the pixel difference histogram while preserving the advantage of low visual distortion of the PVD. The histogram-based steganalysis is therefore defeated.

Steganography with least histogram abnormality

A novel steganographic scheme is proposed which avoids asymmetry inherent in conventional LSB embedding techniques so that abnormality in the image histogram is kept minimum. The proposed technique is capable of resisting the χ 2 test and RS analysis, as well as a new steganalytic method named GPC analysis as introduced in this paper. In the described steganographic technique, a pair of mutually complementary mappings, F 1 and F − 1, is used, leading to a balanced behavior of several statistical parameters explored by several steganalytic schemes, thus improved security. Experimental results are presented to demonstrate the effectiveness of the method.

Analysis of parity assignment steganography in palette images

In parity assignment-based steganography for palette images, all colors in a host image are divided into two subsets, and each pixel is used to carry one secret bit. This paper describes an analytic method against the parity assignment-based steganographic techniques. By finding the rule of color modifications, a steganalyst can attempt to recover the original histogram in a way that is a reverse of data embedding. Because of the abnormal colors in the original image, an excessive operation will cause some negative values in the recovered histogram. This provides a clue for revealing the presence of secret message and estimating the length of embedded bit sequence.

Data Hiding in Digital Audio by Frequency Domain Dithering

A technique that inserts data densely into short frames in a digital audio signal by frequency domain dithering is described. With the proposed method, large embedding capacity can be realized, and the presence of the hidden data is imperceptible. Synchronization in detection is achieved by using a two-step search process that accurately locates a PN sequence-based pilot signal attached to the data during embedding. Except for a few system parameters, no information about the host signal or the embedded data is needed at the receiver. Experimental results show that the method is robust against attacks including AWGN interference and MP3 coding.

Watermarking scheme capable of resisting attacks based on availability of inserter

Attacks based on the presence ofwatermark inserter are easy to perf orm since they make use ofsimilarity between an original watermark and additionally added ones by using the same inserter and key. In this paper, a novel watermarking scheme capable ofresisting inserter attacks is proposed. Watermark signals corresponding to the same key are mutually independent ifthey are randomly selected using the described technique. Thus the inserter attack is invalidated. Performance ofthe proposed method is studied, and simulation experiments presented. ? 2002 Elsevier Science B.V. All rights reserved.

New metric for quality assessment of digital images based on weighted mean square error

In this paper, an image quality measure, termed pixel-based correlation weighted-mean square error (WMSE), is presented. The proposed distortion measure depends not only on the mean square error in the distorted image, but also on correlation between pixels in a predefined neighborhood. Experimental results are given to show the advantage of the described method.

Detection and Removal of Hidden Data in Images Embedded with Quantization Index Modulation

Methods of attack on watermarks embedded with quantization index modulation and block DCT are introduced in this paper. The proposed method can detect the presence of hidden data by analyzing the histogram or spectrum of the transform coefficients, and prevent their extraction without further degrading the image. It is shown that, in some cases, the embedding-induced distortion can even be reduced in terms of peak-signal-to-noise ratio. Experimental results are given to show the effectiveness of the method.

Invertibility attack against watermarking based on forged algorithm and a countermeasure

It is shown in this paper that, even with a non-invertible watermarking algorithm or an asymmetric watermarking protocol, it is still possible to effect an invertibility attack, which relies on a forged watermarking algorithm, a counterfeit mark, and a fake key. Two examples are given to show the vulnerability of the unfortified non-invertible algorithm/asymmetric protocol. As a solution, a secure watermarking protocol is proposed, which establishes correlation between the watermarking algorithm and the embedded mark.

Multispectral image watermarking based on KLT

This paper describes a triple-layered watermarking scheme employing KLT and two layers of discrete cosine transforms, with data shuffled pseudo-randomly prior to the second DCT. In this way, the watermark is imperceptibly and robustly embedded into a set of multispectral images. The watermark is extractable without original host images. Simulation experiments are carried out to study the anti-attack performance. It is shown that the proposed technique is applicable to color and remote sensed multispectral images.

Multi-bit watermarking scheme based on addition of orthogonal sequences

In this paper, a scheme of watermark embedding based on a set of orthogonal binary sequences is introduced. The described technique is intended to be incorporated into various public watermarking frameworks developed for different digital media including images and audio signals. Unlike some previous methods using PN sequences in which each sequence carries only one bit of the watermark data, the proposed approach maps a number of bits to a single sequence from an orthogonal set. Both analytical and experimental studies show that, owing to the full exploitation of information carrying capability of each binary sequence, the performance is significantly improved compared with previous methods based on a one-bit-per-sequence technique.