Steganalysis based on wavelet packet transform and genetic algorithm

作者：Zhou Zhi Ping^*; Zhang Xiao Xiang; Hui Mao Mao

来源：Systems Engineering-Theory & Practice, 2010, 30(10): 1864-1869.

• Characteristic functions
• Classification accuracy
• Cover-image
• Detection accuracy
• Image library
• Original images
• Orthogonal wavelets
• Prediction-error image
• Steganalysis
• Stego image
• Sub-bands
• Universal steganalysis
• Wavelet package decomposition
• Wavelet packet transforms

Against the shortness of general steganalysis method which use the orthogonal wavelet to extract the characteristics, a effective steganalysis method based on wavelet package decomposition (WPD) and genetic algorithm was designed. Firstly, by three scales of WPD, image and its prediction-error version were decomposed into some coefficients subbands, the multi-order characteristic CF) moments of histogram were extracted from these subbands. Then genetic algorithm (GA) was employed to search and identify the potential informative features combinations for classification. The classification accuracy from the support vector machine classifier is used to determine the fitness in genetic algorithm. The experiments works on mixed image library, constituting by Outguess, F5, MB1 stego-images and cover images. Results show the method can detect stego and original images reliably, and the average detection accuracy of this method exceeds those of its closest competitors by at least 1.7% and up to 17.8%.

• 出版日期2010
• 单位江南大学