Coverless Image Steganography Based on Optical Mark Recognition and Machine Learning

Autor: Al Hussien S. Saad, M. S. Mohamed, Eslam H. Hafez
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Zdroj: IEEE Access, Vol 9, Pp 16522-16531 (2021)
Druh dokumentu: article
ISSN: 2169-3536
DOI: 10.1109/ACCESS.2021.3050737
Popis: The most significant factor to consider during private information transmission through the internet (i.e., insecure channel) is security. So, to keep this data from unauthorized access during transmission, steganography is used. Steganography is the scheme of securing sensitive information by concealing it within carriers such as digital images, videos, audio, text, etc. Current image steganography methods work as follows; it assigns cover image then embeds the secret message within it by pixels' modifications, creating the resultant stego-image. These modifications allow steganalysis algorithms to detect the embedded secret message. So, a coverless data hiding concept is proposed to solve this problem. Coverless does not mean that the secret message will be transmitted without using a cover file, or the cover file can be discarded. Instead, the secret message will be embedded by generating a cover file or a secret message mapping. In this paper, a novel, highly robust coverless image steganography method based on optical mark recognition (OMR) and rule-based machine learning (RBML) is proposed.
Databáze: Directory of Open Access Journals