Detection of Steganography Based on Wavelet Statistics and Multi-Class SVM
Yijie Wu · Computer Engineering and Applications Journal · 2006
A new detection of steganography algorithm based on wavelet statistics of color images and multi-class SVM is proposed.In order to capture some regularities which are ignored when converting color images into grayscale images among the color channels,statistical models for color images are built.The wavelet decomposition is implemented in each color channel,the magnitude of decomposition coefficients and the log error between the actual coefficient and the predicted coefficient magnitudes are used to yield statistics.The multi-class support vector machine algorithm has been employed in the pattern discrimination.Stego images created by several tools are tested under certain embed rate.The experimental results show that the algorithm has stronger universal performance and higher discriminating rate.