Improved Security of a Deep Learning-Based Steganography System with Imperceptibility Preservation

Abstract

Since its inception, the steganography system (SS) has continuously evolved and is routinely used for concealing various sensitive data in an imperceptible manner. To attain high performance and a better hiding capacity of the traditional SS, it has become essential to integrate them with diverse modern algorithms, especially those related to artificial intelligence (AI) and deep learning (DL). Based on this fact, we proposed a DL-based SS (DLSS) to extract some significant features (like pixel locations, importance, and proximity to the imperceptibility) from the cover image using a neural network (NN) in a hierarchical form, thus selecting the candidate pixels for embedding afterwards. The pixel weight was expressed in terms of the position, imperceptibility, and its relationship with adjacent pixels to be a stego image. Performance evaluation revealed that the proposed DLSS achieved imperceptibility of 84 dB for images in training mode of a standard dataset.

A Deep Dive into Deep Learning-Powered Steganography for Enhanced Security: Review

“…In response to these issues, DL has developed as a commonly utilized component of ML systems [37]. Figure (3) shows an example of a typical AI architecture, emphasizing how different technologies are integrated to tackle complicated tasks efficiently. In order to extract the embedded secret image using these methods, the cover image and the secret image are fed into an encoder to create a stego image, which is subsequently sent through a decoder.…”

“…Protecting data, ensuring data transmission that respects privacy, and preventing data breaches and plagiarism are the main goals of current research. Steganography systems (SS) are one of the many protection strategies that have been quickly developed in response to the ever-evolving threats posed by adversaries and hackers [3]. Historically, safe data transfer has been achieved through the use of encryption techniques; however, worries about unapproved access and modification have prompted research into other approaches.…”

Addressing the Vulnerability of Data Routing in IoT Network based on Optimization Techniques and Advanced Blow Fish Encryption

“…Presently, The final stage after optimizing the routing path of data in IoT network is securing transmitted data, in this regard the proposed study developed an advanced blowfish encryption technique. The conventional blow fish encryption is considered one of the most dominant encryption techniques in WSN applications [34], [35] , Figure 5 shows the typical configuration of blow fish encryption. In this study, Cipher Block Chaining (CBC) was utilized to improve the typical blow fish technique, sequential blocks of data are chained to disappear the matching blocks of plaintext in the cipher data.…”

“…However, many limitations obstruct sensors` performances including but not limited to energy resources, memory, and computation [12][13][14]. There are numerous studies have been conducted to address these limitations particularly energy efficiency in WSN based on IoT.…”