Lossless Data Hiding Based on Histogram Modification of Difference Images

Abstract

Abstract. In this paper, we propose a new lossless data hiding method where distortion due to data embedding can be completely removed from the watermarked image after the watermark has been extracted. In the proposed method, we utilize characteristics of the difference image and modify pixel values slightly to embed the data. We show that the lower bound of the PSNR (peak-signal-to-noise-ratio) values for typical images are about 51.14 dB. Moreover, the proposed method is quite simple and fast. Experimental results demonstrate that the proposed scheme can embed a large amount of data while keeping high visual quality of test images.

Steganalysis for Reversible Data Hiding

“…After embedding a set of message into a set of original image using DIH method get a set of stego image, we form the difference image D(i,j) and calculate the histogram of D(i,j) for each image in the original image set and the stego image set, we found out that DIH method changed natural of the difference image histogram of typical image significantly as the example in [3].…”

“…In other words, in a stego image, τ time. In certain cases, the sum of h this image is a stego image using DIH method and was embedded in original image as We assume that the to be embedded of bit "1" equals to the number of bit sequence into an original image store n/2 message bits, it means and h -2 , remaining n/2 data bits are embedded into original Lena grayscale image of size 512 x 512 pixels (Figure perform the difference image D(i,j) and calculate histogram of all D(i,j) that is shown in embed a watermark which is a binary logo image of size 128x56 s, equivalent to a binary sequence of 7,168 bits (following the example of [3] DIH. We get the difference image histogram in Figure 1.…”

“…In this subsection, we describe again details of Lee and his colleague's lossless data hiding method using the histogram modification of the difference image [3].…”

“…Lossless data embedding can take place in the spatial domain [1,2,3], or in the transform domain [4,5]. Lee et al [3] proposed a lossless data embedding technique (we assume that the technique name DIH method), which utilizes characteristics of the difference image and modifies pixel values slightly to embed the secret data. Xuan et al [5] proposed a histogram shifting method in integer wavelet transform domain (we assume that the technique name IWH method).…”

Robust Colour Image Watermarking Scheme Based on Feature Points and Image Normalization in DCT Domain

“…However, watermarking methods involving invariant domains are difficult to implement due to the log-polar mapping [17]. Based on the fact, that the histogram is independent of the position of the pixels, the authors in [18,19] presented histogram-based watermarking approaches. However, these approaches suffer from robustness limitations under histogram enhancement and equalization attacks.…”