Binary Image Steganography Analysis Method

Steganography and steganalysis have become hot spots in the field of information security since 1990s. The relationship between Steganography and steganalysis, similar to that of Cryptography and cryptanalysis, both are indispensable, and they promote each other for common development.This paper comprehensively introduces the date hiding at first, and then does research on steganalysis in binary images and steganography in gray-scale images. The main achievements are as follows:(1) Through massive experiments we found a law. The statistical characteristics of run length between cover image and the image after test embedding have obvious difference, while the statistical characteristics of run length between stego image and the image after test embedding is similar for the binary text image embedding algorithm which hides information by flipping center pixel of the image block. Based on this law, a novel technique for the steganalysis of binary text image is presented in this paper. It can detect the existence of a secrtet message hidden by the embedding algorithms which hide information by flipping block center pixel. Through the classification experiment, the threshold of several typical characters had identified and the experimental results showed that detection rate can reach 99.93% by using the proposed method and the related threshold.(2) This paper builds high-scale statistical model for capturing statistical difference between binary cover images and binary stego-images. By exploring analysis of variance, the statistics that are more sensitive to binary stego-images are chosen as features of images. Kernel-based support vector machine is chosen as classifier for binary cover images and binary stego-images. Through the classification experiment, our method can reach blind steganalysis of data hiding in binary images by the embedding algorithms which hide information not by flipping block center pixel.(3)This paper designs a general binary image steganalysis method. It not only can detect data hiding in binary images by the embedding algorithms which hide information by flipping block center pixel successfully, but also can detect data hiding in binary images by the embedding algorithms which hide information not by flipping block center pixel. It is an excellent blind steganography in binary image.(4) A new method for hiding secret information in gray-scale images is proposed. Secret information is hidden in higher bit level of pixels using steganography of binary image. Experiments show that the stego-images have no perceptual abnormality with PSNR values greater than 40 dB when an embedding rate reaches 0.76 bits per pixel, and can resist several effective methods for LSB steganalysis teganalytical detection giving consideration to good trade-off between securities and embedding capacity.