Steganalysis Based on Lifting Wavelet Transform for Palette Images

Abstract

Based on the statistical analysis of the correlation between the low frequency sub-band coefficients and the high frequency sub-band coefficients in wavelet domain, a novel steganalytic algorithm with the ability of reliable detecting modifications in palette images is proposed in this paper. First, we divide the image into non-overlapping 8times8 blocks and perform integer wavelet transform based on lifting scheme to each block. Then, the average correlation, which is used to determine whether the secret message exists or not in palette images, is calculated according to the correlation between low frequency coefficients and high frequency coefficients. Experimental results show that the new algorithm can achieve better performance than the pairs analysis. Moreover, it is easy to be implemented and well-suited for real-time detection.