Chaos-based Spatial Steganography System for Images

Abstract

In this paper, we propose a chaos-based enhancement of two spatial steganographic algorithms; the AE-LSB and the EA-LSBMR and our objective is to study their performances. The first algorithm is an adaptive LSB (Least Significant Bit) steganographic method using pixel value difference that provides a large embedding capacity and imperceptible stego images. The second method is an edge adaptive scheme which can select the embedding region according to the size of secret message and the difference between two adjacent pixels in the cover image. The two methods suffer from low security against attacks that try to recover secret data.To overcome this weakness, we propose to enhance the message security of these methods. The enhancement consists of using an efficient chaotic system in order to choose in a pseudo-chaotic manner the pixels in the cover image where the bits of the secret message will be embedded. In this way, the inserted message becomes secure against message recovery attacks and becomes as well spread over the whole image in a uniform manner. Experiments show that the security of the algorithms is increased.