Deep learning is regarded as a promising solution for reversible steganography. There is an accelerating trend of representing a reversible steo-system by monolithic neural networks, which bypass intermediate operations in traditional pipelines of reversible steganography. This end-to-end paradigm, however, suffers from imperfect reversibility. By contrast, the modular paradigm that incorporates neural networks into modules of traditional pipelines can stably guarantee reversibility with mathematical explainability. Prediction-error modulation is a well-established reversible steganography pipeline for digital images. It consists of a predictive analytics module and a reversible coding module. Given that reversibility is governed independently by the coding module, we narrow our focus to the incorporation of neural networks into the analytics module, which serves the purpose of predicting pixel intensities and a pivotal role in determining capacity and imperceptibility. The objective of this study is to evaluate the impacts of different training configurations upon predictive accuracy of neural networks and provide practical insights. In particular, we investigate how different initialisation strategies for input images may affect the learning process and how different training strategies for dual-layer prediction respond to the problem of distributional shift. Furthermore, we compare steganographic performance of various model architectures with different loss functions.

Trust and security are fundamental to the successful adoption of the Internet of Things (IoT). This paper proposes a secure message authentication scheme based on steganographic secret sharing for building trust in IoT systems. In our scheme, the message is split and distributed to two participants by a dealer, and it can be revealed only when the two authorized participants grant their consents. Neither of the participants can disclose the message without the consent of the other. To avoid malicious cyberattacks in IoT communications, each share of message is concealed in the form of a human face image, referred to as the shadow image, via a generative adversarial network (GAN). For each participant, a convolutional neural network (CNN) is trained to extract the share of message from the shadow image generated with the participant’s key. Distortions and alterations to the shadow images may occur during the transmission from the dealer to the participant. As a tamper-evident design, each shadow image is morphed with the participant’s source image under the participant’s customized morphing parameter. Cheater detection is also crucial for involved participants to identify fake shadow images during the secret retrieval process. As a cheating countermeasure, the shadow image for one participant is morphed with that for the other and then morphed further with the given source image. The proposed scheme enables multi-factor authentication in the sense that the message is protected by the keys, source images, and morphing parameters of two participants.

Wireless ad hoc networks, especially in the hostile environment, are vulnerable to traffic analysis which allows the adversary to trace the routing messages and the sensitive data packets. Anonymity mechanism in ad hoc networks is a critical securing measure method employed to mitigate these problems. In this paper, we propose a novel secure and anonymous source routing protocol, called SADSR, based on Dynamic Source Routing (DSR) for wireless ad hoc networks. In the proposed scheme, we use the pseudonym, pseudonym based cryptography and the bloom filter to establish secure and anonymous routing in wireless ad hoc networks. Compared to other anonymous routing protocol, SADSR is not only anonymous but also the secure in the routing discover process and data transmission process.

This paper presents an improved secure reversible data hiding scheme in encrypted images based on integer transformation, which does not need using a data hider key to protect the embedded secret data. We first segment the original image into blocks of various sizes based on the quadtree-based image partition. For each block, we reserve m least significant bits (LSBs) of each pixel as embedding room based on the reversible integer transformation. In order to improve the security of the image encryption, we pad the mLSBs of each pixel using the corresponding (8-m) most significant bits (MSBs) information after the transformation, which protects the security of the encryption key. Then, we encrypt the transformed image with a standard stream cipher. After the image encryption, the data hider embeds the secret data in the mLSBs of the encrypted images through an exclusive or operation. On the receiving side, the receiver can extract the secret data after the image decryption and recover the original image without loss of quality. The security analysis shows that the proposed scheme improves the security weakness of the scheme directly using adaptive integer transformation. The experimental results show that the proposed method achieves a higher embedding ratio compared with several relevant methods.

Secret image sharing is a hot issue in the research field of data hiding schemes for digital images. This paper proposes a general $\left(k,n\right)$ threshold secret image sharing scheme, which distributes secret data into $n$ meaningful image shadows based on a non-full rank linear model. The image shadows are indistinguishable from their corresponding distinct cover images. Any $k$ combination of the $n$ shares can perfectly restore the secret data. In the proposed scheme, the integer parameters $\left(k,n\right)$, with $k\le n$, can be set arbitrarily to meet the application requirement. The experimental results demonstrate the applicability of the proposed general scheme. The embedding capacity, the visual quality of image shadows, and the security level are satisfactory.

Edge Detection Based on Fuzzy Logic and Expert System | InTechOpen, Published on: 2012-05-09. Authors: Shuliang Sun, Chenglian Liu, Sisheng Chen