Estimating PQ Algorithm Message Length Using First- and Second-Order Statistics

Steganography, the technology of protecting a secret message by embedding it inside a cover image, continues to be investigated and enhanced as an alternative data protection method. This paper deals with hiding multimedia files in true color RGB cover images with an emphasis on reducing the cover size, increasing hiding capacity and enhancing security of the hidden data. A proposed model (DuoHide) is presented in which a secret multimedia file, regardless of its type, is processed without un-compression, and divided between two cover images of equal size and dimensions. The multimedia file is read as a stream of bytes and split vertically into two parts, one part contains the least significant half-bytes, and the other part contains the most significant half-bytes. The two parts are hidden inside two uncompressed RGB cover images using a least significant 4-bit replacement technique. The resulting dual stego images are expected to be sent separately, through different channels, to avoid capture of both stego files by an adversary. Extraction of the secret file is achieved through merging LSB half-bytes from the two stego files. The extracted file is identical in content and structure with the original secret file. The implemented DuoHide system was evaluated using a set of public multimedia files, images, audios, and videos, of various sizes. The secret file sizes ranged from 5% to about 100% of the cover image's size. The experimental results showed that even at the highest embedding ratio, which is based on the secret-to-cover ratio, there were no perceptible visual differences between cover and stego images. The PNSR value was calculated as PSNR1, for cover1 and stego1, and PSNR2 for cover2 and stego2. The lowest PSNR value was around 31 dB for the highest embedding ratio, which is considered acceptable concerning statistical imperceptibility. The PSNR value increased as the embedding ratio decreased, reaching around 65 Decibel (dB) for the case of 5% secret-to-cover ratio. The integrity of the extracted secret file was verified through a bitwise comparison between original and extracted files, which showed zero differences. The DuoHide model is expected to provide better security for the hidden file, in case an attacker manages to capture one of the stego images and recover the hidden content because the attacker will only get an incomprehensible set of half-byte bits. An additional advantage of using a pair of stego files is that of reducing stego file size by 50%, to avoid problems and limitations of transmitting large files, especially that multimedia files are often large, and they cannot be compressed because they are already compressed. Security of the DuoHide system can further be improved by randomizing storage locations within the two stego images.

The multi-bit image steganography (Paul et al. Multimedia Tools and Applications 76(5):7445–7471, 2017) proposed by Paul et al. provides high embedding capacity, and adapts the embedding position of messages to image content. This paper investigates the security and robustness of multi-bit steganography. In detection resistant experiment, we embed secret messages into cover images with multi-bit image steganography to get corresponding stego images, and adopt a more popular steganalyzer (SRM) to extract features of cover images and stego images. The ensemble classifier is trained with default settings. The experimental results show detection resistance performance of multi-bit image steganography is poor. In robustness test, we embed secret messages into cover images with multi-bit image steganography to get stego images, and attack the stego images with Stirmark. Then, we extract the secret messages from the attacked stego images. The experimental results demonstrate that we can not recover the secret messages from the attacked stego images. Therefore, there are flaws in detection resistant test and robustness test of multi-bit image steganography.

Data is an asset for all parties. So that it must always be guarded or secured so that it is not misused by irresponsible parties. One of the methods used for data security is steganography. In this paper, efforts to secure data are discussed through the development of steganographic algorithms using the Haar integer wavelet transform (IWT-Haar) and least significant bit (LSB) techniques to hide secret messages in the form of digital text into digital images (cover images) to produce stego images. The results show that the extracting process is much faster than the embedding process. The PSNR values obtained ranged from 55-71,5 dB (far greater than 40 dB). This shows that the resulting stego image is of very good quality. In plain view, the difference between the cover image and the stego image has not changed significantly.Key words: digital image, embedding, extraction, Haar integer wavelet transform, steganographyÂ

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

R261, Jana et al. proposed the partial reversible data hiding (PRDH) based on (7, 4) Hamming code, which deals with three images: the original image (OI), the cover image (CI), and the stego image (SI). The CI is obtained by slightly modifying OI. After that, one can embed secret into CI to generate SI. The so-called partial reversible feature is that we can reconstruct the CI faultlessly, which is almost the same to the OI. However, in Jana et al.'s PRDH, the authors only adjust redundant bits by using (7, 4) Hamming code with even (or odd) parity. This simple approach may modify 3 redundant bits at most in 7 least significant bit (LSBs) to obtain CI from OI, and this degrades the CI quality seriously. In this paper, we construct two schemes: the proposed PRDH and the modified PRDH. The proposed PRDH enhances the partial reversible property that improves the visual quality of CI. And, the modified PRDH enhances the proposed PRDH to achieve the high embedding capacity. In addition, we also point out two inaccurate descriptions in Jana et al.'s PRDH about position of embedded secret to more simplify the procedure of embedding secret.

Binary steganalysis is a prediction of whether an image is a cover image (does not contain a secret message) or a stego image (contains a secret message). CNNs are now widely used to find hidden messages embedded in images. CNN optimization strategies such as the use of Spatial Rich Model (SRM) Filter and Batch Normalization, Spatial Dropout layer, and three Fully Connected Layers before softmax have shown improved accuracy on three binary steganalysis CNNs namely Xu Net, Ye Net, and Yedroudj Net as well as two CNNs for image classification (VGG-16 and VGG-19), but which computational elements are relevant to the accuracy of binary steganalysis CNNs remains unknown. Therefore, this study will investigate the factors that affect the performance of CNNs of binary steganalysis that have not been tested before, such as GBRAS Net and Ntivuguruzwa Net. The data set used in this research is Bossbase 1.01 inserted with secret messages using the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$S$</tex> Uniward steganography algorithm with a message length of 0.4 bpp. GBRAS Net and Ntivuguruzwa Net have three main parts, namely the preprocessing module, the convolution module, and the classification module. The test results show that SRM filter and batch normalization play an important role in accuracy, but the addition of spatial dropout layers and three fully connected layers in GBRAS Net and Ntivuguruzwa Net does not always provide an increase in accuracy.

This paper presents a novel reversible data hiding method, which can recover the original image content without distortion from the stego images after the secret data have been extracted. The method utilizes the original cover image to finally generate two visually similar stego images by embedding additional data. During the data embedding process, the pixels in the first stego image are utilized to embed secret data by applying the proposed prefix matching technique, which exploits each of the gray values to carry one or two bits only with ± 1 operation. Pixels in the second stego image are adaptively modified to carry rest secret data by referring to the first stego image using the directed LSB embedding. For the recipient, the additional data can be extracted from the two stego images with the DH key, and the original cover image can be recovered with the location map. If the cover image has no boundary gray-values, one can always recover the cover image from the two stego images without the aid of any extra information. Experiments show that the proposed method can achieve desirable embedding capacity and maintain a high level of image quality.

This paper presents a new secret sharing scheme for identification of cheaters in the elections. In a proposed scheme secret image (voter ID) is embedded into cover image (Indian emblem) of same size and generates n stego images. The resulting stego images are visually good and look like a cover image. Various image quality metrics are used to assess the quality of stego images. Out of n stego images only last two stego images are required to reconstruct the secret and cover images without any loss. The proposed scheme offers high embedding capacity. Experimental results of proposed scheme shown to be superior in terms of PSNR and visual quality than existing schemes.

In this study, the StegoCrypt technique is proposed using a combination of Discrete Wavelet Transform (DWT) and One-Time Pad (OTP). Cover image with size 512 ∗ 512 transformed with Wavelet transformation of four levels. For the first level to the third level, the subband LL is selected to obtain LL3 subband. At fourth level, the LL3 is then transformed into HH4 subband with help of wavelet transformation. This is done to gain strength and imperceptibility to the stego image. The secret message that is used is a binary image with a size of 32 ∗ 32. Secret image is encrypted with OTP before it is inserted in the host image. To test the quality of imperceptibility, stego images were measured using PSNR and MSE. While the quality of secret image reconstruction of extraction and decryption results are measured by using NCC. Robustness of stego image is also tested with JPEG compression attacks. From the test results can be concluded that the proposed method works well and get a better quality stego image. The secret image reconstruction results are also perfect as well as robust to JPEG compression attacks.

Steganography is a process to conceal the existence of a message. Image Steganography involves hiding the message onto a cover image to obtain a resultant stego image. This paper discusses a procedure for hiding coloured message images onto coloured cover images using a lossy Image Steganography technique. The secret images undergo Discrete Cosine Transformation, quantization and AES encryption before embedding using the Least Significant Bit embedding (LSB) technique onto the cover image. Finally, Peak signal-to-noise ratio (PSNR), Structural Similarity Index measurement (SSIM), zero normalized cross-correlation (ZNCC) are used for comparing the cover and stego images as well as the extracted and original images.

Internet proliferation and technological progress have made multimedia information quickly accessible, but they have also posed a threat to privacy and security. Researchers have been interested in digital images due to their capacity to store large amounts of data due to the possibility of protecting sensitive information through digital steganography. Despite their visual imperceptibility, robustness, and ability to embed information, existing image steganography techniques face several challenges. To overcome these challenges, a novel image steganography approach based on a blind model strategy has been proposed for hiding covert messages. The model consists of two stages: embedding and extracting. In the embedding stage, a suitable cover image is selected using the FAST feature point detector. A text message is then converted to a QR code and embedded in the feature points' neighbors using knight tour steps in a chess game. The result is a stego image that appears identical to the original cover image but contains the secret message in its feature points' neighbors. The extracting stage involves finding the feature points and extracting the QR code to obtain the original text message. Since the feature points were not altered during the embedding process, the proposed model is known as a blind model. This approach eliminates the need for the original cover image during the extracting stage. The proposed model was evaluated using several metrics, including the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). The results demonstrate that the proposed algorithm can effectively embed and extract secret messages with high accuracy while maintaining the visual quality of the cover image with 100% of (SSIM), and 73.48 as an average of (PSNR).

Recent studies have demonstrated that a well designed deep convolutional neural network (CNN) model achieves competitive performances on detecting the presence of secret message in digital images, compared with the classical rich model based steganalysis. In this paper, we propose to investigate a category of very deep CNN model–the deep residual network (DRN), for steganalysis. DRN is suitable for steganalysis from two aspects. For the first, the DRN model usually contains a large number of network layers, which proves to be effective to capture the complex statistics of digital images. For the second, DRN's residual learning (ResL) method actively strengthens the signal coming from secret messages, which is extremely beneficial for the discrimination between cover images and stego images. Comprehensive experiments on standard dataset show that the DRN model achieves very low detection error rates for the state of arts steganographic algorithms. It also outperforms the classical rich model method and several recently proposed CNN based methods.

A novel algorithm for colour image steganography using a new intelligent technique based on three phases

This paper proposes an active steganalysis method for quantization index modulation (QIM)-based steganography. The proposed nonparametric steganalysis method uses irregularity (or randomness) in the test image to distinguish between the cover image and the stego image. We have shown that plain quantization (quantization without message embedding) induces regularity in the resulting quantized object, whereas message embedding using QIM increases irregularity in the resulting QIM-stego. Approximate entropy, an algorithmic entropy measure, is used to quantify irregularity in the test image. The QIM-stego image is then analyzed to estimate secret message length. To this end, the QIM codebook is estimated from the QIM-stego image using first-order statistics of the image coefficients in the embedding domain. The estimated codebook is then used to estimate secret message. Simulation results show that the proposed scheme can successfully estimate the hidden message from the QIM-stego with very low decoding error probability. For a given cover object the decoding error probability depends on embedding rate and decreases monotonically, approaching zero as the embedding rate approaches one.

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.