Chih-Wei Shiu

Reversible data hiding for high quality images using modification of prediction errors

In this paper, a reversible data hiding scheme based on modification of prediction errors (MPE) is proposed. For the existing histogram-shifting based reversible data hiding techniques, though the distortion caused by embedding is low, the embedding capacity is limited by the frequency of the most frequent pixel. To remedy this problem, the proposed method modifies the histogram of prediction errors to prepare vacant positions for data embedding. The PSNR of the stego image produced by MPE is guaranteed to be above 48dB, while the embedding capacity is, on average, almost five times higher than that of the well-known Ni et al. techniques with the same PSNR. Besides, MPE not only has the capability to control the capacity-PSNR, where fewer data bits need less error modification, but also can be applied to images with flat histogram. Experimental results indicate that MPE, which innovatively exploits the modification of prediction errors, outperforms the prior works not only in terms of larger payload, but also in terms of stego image quality.

A high capacity reversible data hiding scheme using orthogonal projection and prediction error modification

The payload of a histogram-shifting based reversible data hiding technique is primarily determined by the peak height of the corresponding image histogram. It is known that the peak height of the prediction error histogram is usually higher than that of image histogram itself. In 2009, Tsai et al. adopted the idea of histogram-shifting technique and embedded message by modifying the prediction errors to achieve a higher payload. They used a block-based approach and leaved some pixel values unmodified to predict other pixel values in a block. However, the unmodified pixels took up some embeddable space, causing a reduction in payload. In this paper, a new reversible data hiding technique based on prediction error modification is proposed. The orthogonal projection technique is employed to estimate the optimal weights of a linear predictor to increase the prediction accuracy and the embedding capacity. The experimental results revealed that the proposed method outperformed the existing histogram-shifting based reversible data hiding technique in terms of payload and image quality.

A Minimal Euclidean Distance Searching Technique for Sudoku Steganography

Sudoku, a simple and fun game of logic, has been used for steganography to conceal messages into a digital image recently. Chang et al. adapted the idea of smallest Manhattan distance, embedding secret messages into the neighbors of the located element according to a given Sudoku solution. Hong et al. improved Chang et al.psilas technique by introducing additional set of candidate elements to reduce distortions. However, the aforementioned methods suffer from undesirable distortions because the Manhattan distance architectures are used in their method. The proposed method suggests a new scheme for searching embedding positions based on the nearest Euclidean distance, so that minimal distortions can be reached. The experimental results show that, in average, the visual quality of stego image is 1.70 dB higher than that of Chang et al.psilas method, and 0.72 dB higher than that of Hong et al.psilas method under the same embedding capacity.

Steganography for BTC compressed images using no distortion technique

Abstract In this paper, we proposed an improved data hiding scheme to embed secret data in the compressed bitstreams where the quality of the image is maintained even after embedment. The difference in the quantized values for each block is used to determine whether only 1 bit of secret is to be hidden for the block or to toggle bits in the bitmap to hide more bits. The scheme has a high payload and is a no distortion data hiding. Experimental tests with Lena showed that in the 2 × 2 block-sized BTC compressed bitstreams, an additional 10 488 bits were embedded and 416 bits for the 4 × 4 block-sized. The new improved scheme increases payload and is reversible.

Lossless Steganography for AMBTC-Compressed Images

A lossless or reversible steganography is defined as an original image can be completely recovered from the stego-image after the embedded data has been extracted. This technique has been focused on spatial uncompressed domain recently, and is considered more challenging to carry out in the compressed domain. In this paper, we propose a lossless, compressed domain steganography technique for AMBTC-compressed images based on the interchange of two quantization levels. The location map, which is often used in lossless steganography, is not required in our scheme. Therefore, the hiding capacity is independent of the compressed codes. Moreover, the stego-image preserves the same image quality as the original compressed images. The experimental results show that the proposed scheme is not only easy to implement but also provides an efficient mechanism for lossless data embedding.

Steganography Using Sudoku Revisited

In this paper, we propose a revised version of information hiding scheme using Sudoku. The original work was proposed by Chang et al. in 2008, and their work was inspired by Zhang and Wangs method and Sudoku solutions. Chang et al. successfully used Sudoku solutions to guide cover pixels to modify pixel values so that secret messages can be embedded. However, all possible embedding positions for smaller image distortions in Chang et al.s method are not fully explored. The proposed method modified Chang et al.s embedding rules by allowing candidate pairs falling into more suitable regions in the reference matrix, so that a better position for data embedding can be selected to minimize image distortions. The experimental results show that, in average, the visual quality of stego image is 0.81 dB higher than that of Chang et al.s method without sacrificing any embedding capacity.

An improvement of diamond encoding using characteristic value positioning and modulus function

The diamond encoding technique controls payload and image quality by a k value. Although a small increase in k can increase payload, it also increases image distortion. In this paper, we proposed an improvement scheme to the diamond encoding technique for reducing image distortion during a k change. Another aim is to increase payload size. A square matrix conversion of the diamond matrix is used to lower the MSE values in high payload when k=3. A lower MSE reduces image distortion. In low payload, that is when k changes from 1 to 2, a one dimension matrix modulus is used to reduce image distortion. Experimental results showed that payload may be increased while image quality is not significantly reduced.

A High Quality Histogram Shifting Based Embedding Technique for Reversible Data Hiding

Reversible data hiding is a technique that embeds secret information into a host media without loss of host information. Ni et al.s histogram shifting technique is a high-quality, reversible method for data embedding. However, their technique still suffers from undesirable distortion at low embedding rates and lack of a mechanism to control the stego-image quality, due to all pixels between the peak point and the minimum point have to be shifted one unit for data embedding. The proposed technique lowers the distortion performance at low embedding rates by scanning pixels from outer region toward the center of the host image, and chooses better locations for shifting histograms to embed data. The experimental results show that the proposed method significantly improves the quality of the stego image of histogram shifting technique, especially at low embedding rates.

A New Reference Pixel Prediction for Reversible Data Hiding with Reduced Location Map

In this paper, a new reversible data hiding method based on a dual binary tree of embedding levels is proposed. Four neighborhood pixels in the upper, below, left and right of each pixel are used as reference pixels to estimate local complexity for deciding embeddable and non-embeddable pixels. The proposed method does not need to record pixels that might cause underflow, overflow or unsuitable for embedment. This can reduce the size of location map and release more space for payload. Experimental results show that the proposed method is more effective in increasing payload and improving image quality than some recently proposed methods.

A secure data hiding method based on preservation of saturated pixels

Zhang and Wang in 2007 proposed an exploring the modification of embedding directions (EMD) method for which n pixels are used as an embedding unit, and a digit in based 2n+1 can be concealed by modifying a pixel value a grayscale at most. In case the overflow or underflow problems occur, EMD adds or subtracts one of the saturated pixels in the same pixel group by one and re-embeds the digit. However, if the number of saturated pixels is considerable, the modification of the saturated pixels may result in a critical vulnerability to the LSB-based steganalyzers. This paper proposes a method to leave those saturated pixels untouched and embeds only those non-saturated ones. Because no saturated pixels are modified, the stego image is insensitive to the detection by the LSB-based steganalyzers. The experimental results reveal that the proposed method is more robust to the LSB-based steganalyzers than the original EMD method when the number of saturated pixels is considerable.