Texture-Image-Oriented Coverless Data Hiding Based on Two-Dimensional Fractional Brownian Motion

• DOI: 10.3390/electronics13204013
• Publication Date: 2024-10-14T11:47:05Z
• Authors: Yen-Ching Chang; Jui-Chuan Liu; Ching-Chun Chang; Chin-Chen Chang
• Keywords: Fractional Brownian motion; Texture (cosmology)

In an AI-immersing age, scholars look for new possibilities of employing AI technology to their fields, and how strengthen security protect privacy is no exception. a coverless data hiding domain, the embedding capacity image generally depends on size chosen database. Therefore, choosing suitable database critical issue in hiding. A novel approach proposed by applying deep learning models generate texture-like cover images or code images. These are then used construct steganographic transmit covert messages. Effective mapping tables between hash sequences established during process. The generated two-dimensional fractional Brownian motion (2D FBM) simply called (FBIs). only parameter, Hurst exponent, 2D FBM determines patterns these images, seeds random number generator determine various appearances pattern. Through FBM, we can easily as many FBIs multifarious sizes, patterns, possible whenever wherever. paper, model treated secret key selecting qualified encode corresponding sequences. Both different pick out diverse FBIs. scheme effective when amount limited. experimental results show that our more reliable, efficient, higher capacity, compared other methods.