A JPEG-based statistically invisible steganography

The objective of steganography is to implement covert communication by hiding data in digital files such that there is no indication of the existence of hidden data; so an ideal steganographic system should be able to withstand all steganalysis methods--or be undetectable. In image steganography, several information-hiding systems have been proposed for JPEG images. However, with the advances in steganalysis, it seems that fewer and fewer steganography systems are truly undetectable. In this paper, we design a new steganography called statistically invisible steganography (SIS) to hide data in JPEG images, which are invulnerable to current methods of steganalysis. Specifically, we embed data in the selective coefficients of the selective DCT blocks that have high complexity and maintain original statistics. Experimental results show that current techniques of steganalysis are unable to successfully detect SIS steganograms.