DCT based Image Steganographic Method Resisting Statistical Attacks
The aim of steganography is to conceal the very existence of hidden communication, so its demand on security is serious. The security includes both imperceptibility and undetectability. But most steganographic methods didn’t pay enough attention to the undetectability. In this paper, we propose a novel DCT-based steganographic method for images. The method takes advantage of the similarities of DCT coefficients between the adjacent image blocks and makes the embedding distortion spread to the adjacent image blocks. Experimental results demonstrate that this proposed method can not only preserve good image quality, but also resist some typical statistical attacks.
The purpose of steganography is to hide the very presence of communication by embedding messages into innocuous-looking cover objects, such as digital images. To accommodate a secret message, the original cover image is slightly modified by the embedding algorithm to obtain the stego image.
The early steganographic schemes focused on introducing as little distortion in the cover image as possible utilizing the seemingly intuitive heuristics that the smaller the embedding distortion is, the more secure the steganographic scheme becomes. The Least Significant Bit embedding (LSB) with sequential or random message spread is a representative method of the idea, some famous steganographic software, such as Steganos, S-tools and Hide4PGP[1], are all based on the idea of LSB. In addition, researchers proposed some improved LSB replacement me thods[2,3], based on characteristics of human visual systems (HVS) and spatial complexity measure, in order to increase the capacity. However, recent advances in steganalysis clearly showed that this is not the case. The LSB method has been successfully attacked even for very short messages [4,5]. In essence, the LSB embedding is so easily detectable because it introduces distortion that never naturally occurs to images and creates an imbalance between appropriately defined statistical quantities. To enhance the security and improve the undetectability of steganographic method, some researchers attempt to design new steganographic schemes that embed messages by adding Gaussian noise to the image. Lisa M. Marvel describes a high-capacity method for embedding message bits in uncompressed raw image formats [6]. The hidden messages are modulated to Gaussian signals adding to the cover image. Faisal Alturki’s approach [7] is a simple bit-replacement of quantized DCT coefficients calculated from a randomly permuted image. In the method, the distortion distribution is approximately a generalized Gaussian distribution. With the development of the steganalysis, a new steganalytic method has been proposed by Jeremiah J. Harmsena in 2003 [8], which is based on the fact that noise adding in the spatial domain correspondsto low-pass filtering of the histogram. Thus, the histogram of stego images has less power in high frequencies than the same histogram for cover images. Based on this statistical property, a classifier can be established to distinguish between cover images and stego images
Free download research paper
CSE PROJECTS