A Multiple-Chaotic Approach for Steganography

Abstract

In a recent work, chaos has been utilized to modify addresses of message bits while hidden in a cover image. In this study, we extend the above technique to include multiple chaotic maps for increased security. Three systems have been modified using chaotic-address mapping for image steganography in the spatial domain. The first system, the well-known LSB technique, is based on the selection of pixels and then hides secret message in the Least Significant Bits LSBs of the given pixel. The second system is based on searching for the identical bits between the secret message and the cover image. The third system is based on the concept of LSB substitution. It employs mapping of secret data bits onto the cover pixel bits. To increase the security performance of the above chaos-based steganographic techniques, multiple-chaotic maps are introduced in this study by using multiple formulas to generate chaotic sequences used to track the addresses of shuffled bits. The generated chaotic sequences were evaluated to determine the randomness (using correlation tests) and the chaotic characteristics of a nonlinear system (using Lyapunov exponent, Poincaré section and 0-1 test). The performance and security levels of the proposed techniques were evaluated by using Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), histogram analysis and correlative analysis. The results show that the proposed method performs existing systems.