A methodological and analytical framework for image steganography using evolutionary algorithms

Abstract Steganography has been developed with the aim of establishing secure communication in a completely imperceptible manner. In a steganography system, confidential information is embedded within a carrier medium such as an image, audio, or video in such a way that no discernible change occurs in the apparent content of the medium, and an observer or unauthorized attacker is unable to detect the presence of the hidden message. However, the use of detection algorithms can enable the identification of the hidden message; if the detection rate of the message’s presence exceeds the random guess threshold, the steganography system is practically considered insecure. Information steganography is an approach for transmitting confidential data in the form of a cover object that provides the highest level of security, such that even if unauthorized access to the transmitted medium occurs, it is impossible to extract or prove the existence of the hidden data. From this perspective, the fundamental goal of steganography is not merely content confidentiality but the non-provability of the existence of the embedded message. In recent years, numerous methods for image steganography have been proposed, each vulnerable to various attacks and methods for detecting hidden messages. Accordingly, in this paper, the concepts and fundamentals of steganography are first reviewed, and then the prominent methods in the field of image steganography, with a special focus on approaches based on evolutionary algorithms, are examined from an analytical perspective.