Dynamic colour text steganography model using rgb coding and character spacing to improve capacity, invisibility and security

• Main Author: Thabit, Reema Ahmed Abdalla
• Format: Thesis
• Language: English
• Published: 2022
• Subjects: Data protection; Coding theory
• Online Access: http://psasir.upm.edu.my/id/eprint/104008/1/FSKTM%202022%205%20IR.pdf; http://psasir.upm.edu.my/id/eprint/104008/
• Institution: Universiti Putra Malaysia

Protecting sensitive information transmitted via public channels is a significant issue faced by governments, militaries, organizations, and individuals. Steganography protects the secret information by concealing it in a transferred object such as video, audio, image, and text. Text is an ideal object for steganography as it uses low bandwidth and is commonly used. Exploit of text features such as font attributes in text steganography has been proposed, however, many existing feature-based text steganography methods suffer from low capacity, weak invisibility and poor security: low capacity is caused by embedding fewer bits per location, utilizing less usable characters, and low compression efficiency; weak invisibility is due to increased colour differences between the cover and the stego text; while poor security resulted from constant mapping and the permanent sequence selection of embedding positions for the hidden message and stego key. To overcome these problems this study proposed the Colour-spacing Text Stego (DCTS) model, which includes four new techniques: a Secret-block & Colour-spacing Matrices Generation (SCMG) technique to achieve high capacity; the Colour Spacing Normalization (CSN) technique to enhance invisibility; and proposed two techniques for two security layers, i.e. the first security layer, the Dynamic Selection of Embedding Positions (DSEP) technique, which hides the secret message and stego key in dynamic positions; and the second security layer, the Dynamic Colour Spacing Mapping (DCSM), which maps the secret message change dynamically. The results of the study found that the DCTS model produces better performance with a high capacity of 98.85% in a small used space by 5.79%, as well as increases the bits per location by 16 bits. Also, it maintains high invisibility by 5.07% when applying black or coloured cover text. With two security layers, the proposed DCTS achieves high security compared to the existing methods. To conclude, the Dynamic Colour-spacing Text Stego-model (DCTS) embeds a high secret data capacity while maintaining invisibility and security. DCTS model offers a new perspective on feature-based text steganography to protect against visual and statistical attack issues.