Detection of BPCS-steganography using SMWCF steganalysis and SVM

In this paper an improvement to the steganalysis based on statistical moments of wavelet characteristic function (SMWCF) and artificial neural network (ANN) as classifier is presented, previous experiments have showed that this steganalysis system has a good performance in the detection of stego-image created by different steganography algorithms, but it has problems to the steganography based on bit plane complex segmentation (BPCS), this steganalysis has showed a low detection rate of stego-image generated by BPCS steganography, therefore this work proposes to use a support vector machine (SVM) as classifier instead of ANN. Experimental results show considerably increase of BPCS detection rate (more than 20%) when SVM is used, instead of ANN.