Research On Speech Steganalysis For Multiplicative Embedding Model

Steganalysis is a new technology of information security. The primary task of steganalysis is to make statistical analysis of the multimedia signals and determine whether secret information has been hidden in. In this paper, a novel audio steganalysis method based on multiplicative embedding model is proposed. In time domain and DCT domain, the test audio signal is firstly calculated its absolute value and logarithm. Then statistical features of the co-occurrence matrix are selected as classification features. The statistical features are extracted from speech signals and classification method is used to classify the features. In wavelet domain, the test audio signal is firstly calculated its absolute value and logarithm. Then statistical moments of the histogram and the frequency domain histogram, higher-order statistical moments and the co-occurrence matrix are selected as classification features respectively. At last, support vector machine (SVM) is utilized as a classifier to classify the features. Simulation results show that the performance of the proposed scheme is good for multiplicative embedding model.