摘要： 基于精确描述图像小波系数间统计特性的小波域二维隐马尔可夫模型(HMM)参数集合，提出一种针对小波域信息隐藏算法的新型隐写分析技术。通过使用二维HMM对小波系数进行建模，对生成的HMT森林在隐写前后的参数集合构造隐写分类特征，采用SVM分类器进行隐写判别。实验表明该方法适用于小波域隐写术的检测，对小波域QIM、MFP和BPCS隐写有较好的检测性能。

关键词: 小波域隐写分析, 隐马尔可夫模型, 参数集合, 支持向量机

Abstract: A novel steganalysis method is proposed on the basis of 2-D Hidden Markov Model(HMM) in wavelet domain which is employed to describe the statistics of wavelet coefficients precisely. By modeling wavelet coefficient with 2-D wavelet HMM, classification features are constructed based on parameter sets of HMT forests. Experiments show the technology is applicable for the detection of wavelet domain steganography, especially with higher detecting performance for QIM, MFP, BPCS steganography in wavelet domain.

Key words: wavelet domain steganalysis, Hidden Markov Model(HMM), parameter sets, Support Vector Machine(SVM)

中图分类号: 

• TP311