Research On Speech Steganalysis

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. A novel method of wavelet domain speech steganalysis based on statistical features of the co-occurrence matrix is proposed. Firstly, the statistical features which was extracted from time domain speech signals and the statistical features which was extracted from wavelet subbands of speech signals are simulated and finally select the co-occurrence matrix of wavelet subbands as classification features, then feature selection method is used to reduce the dimension of classification features and classification method is used to classify the features. The simulation and analysis of the classification features and classification methods are performed. The detection scheme of wavelet domain speech steganography is constructed based on the simulation results. Then this scheme is utilized to detect the stego-speech signals embedded by wavelet domain LSB, quantization index method and addition method. Simulation results show that the proposed scheme not only reduces the dimension of the feature vector effectively, simplifies the design of the classifier, but also improves the detection performance and the performance of the proposed scheme is good.