Доказательство подлинности фотоизображений встраиванием цифровых водяных знаков
The task of the data protecting by its steganographic hiding in a graphic container is considered. The analysis of method of Kutter-Jordan-Bossen for steganographic embedding of data into a spatial domain of the bitmap image is made. A method for improving some characteristics of Kutter-Jordan-Bossen method is under consideration. The comparison analysis is made. The results of investigation of the improved method are shown in the developed environment of its program realization.
In this paper are shown methods of information protection and also algorithms of embedding of additional information into digital graphics images.
Steganographic concealment of information in digital images frequency domain involves such problems as loss of embedded information and distortion of the natural model of digital images. Previously, the authors of this study proposed an approach to solving the first problem, consisting in iterative correction of embedding errors, and obtained an algorithm based on the QIM method and ensuring the error-free extracting of embedded information through the proposed approach. In this paper, we propose an approach to minimizing the distortions of the natural digital image model in the DCT domain, based on changing the quantization step depending on the image local characteristics in the frequency domain. The result of the work is an improved algorithm for embedding information in the DCT domain of digital images, combining the two proposed approaches. The results of the experiments show that the developed algorithm along with ensuring the error-free extraction makes it possible to reduce the distortions introduced into the DCT coefficients and to ensure the statistical indistinguishability of the cover images and stego images in 75% of cases.
An approach to the detection of hidden information (stegocontainers) in the audio data of MP3 files based on neural network modeling is considered. A multilayer perceptron is used as the instrumental model of the neural network. The structural components of the MP3 file are analyzed: fields containing related information (song titles, album, information about the author, lyrics, etc.), and frames, and fragmented sets of encoded audio data. Useful data are highlighted. A procedure is proposed for presenting audio data of any MP3 file as a uniform set of features of a relatively small size. The dimension of the feature set (data set) can be selected from the range [100-520], in accordance with the minimum and maximum frame size, depending on the compression quality of a single audio file when encoded in MP3 format. Modern software packages for encrypting and decrypting stegocontainers into MP3 files are being investigated. Based on selected software implementations, a database of examples (data sets) is formed from pre-processed MP3 files both containing the stegocontainer and without the stegocontainer. The structure of the neural network for steganalysis of MP3 files is determined experimentally, it is trained and tested. The test results of the neural network system allow us to state its high efficiency
The research of informational content of features in the spatial domain and the frequency domain of compressed JPEG-images at the stegoanalisys is presented. Selection of features with application of the greedy algorithm with an exception is made and the set of informative features on the basis of which criterion function of steganographic embedding is defined is received. Embedding is carried out by means of the algorithm on the basis of replacement operation. Knowing what features contain information about existence of the secret message in the digital image, it is possible to seek for minimization of distortions of these parameters for the purpose of maximizing obscurity of embedding as visually, and for stegoanalyzer. Experiments have shown that such approach allows when embedding more than 4000 bits into the grayscale image with size of 256×256 pixels to reach the minimum distortions of the image, holding a metrics of visual quality PSNR not lower than 52 dB, and stability before the stegoanalyzer.