摘要
为了进一步提高隐写分析算法的检测精度,提出了一种基于成对采样选择性集成的隐写分析算法。从集成分类的特点和隐写分析的特殊性出发,分析了类内以及类间样本之间的联系,研究了4种不同的采样策略,并基于成对采样策略构建选择性集成分类器用于隐写分析。实验表明,不同采样策略能不同程度地影响隐写分析的检测性能;与现有隐写分析方法相比,本文算法能明显降低隐写分析系统的检测错误率(BER)。
A novel steganalysis algorithm based on paired sampling and selective ensemble is proposed to improve detection accuracy of steganalysis. Considering the characteristics of ensemble classification and the particularity of steganalysis, we analyze the relation between two classes of samples and the link a- mong samples of the same class, put forward four different sampling strategies, and build selective en- semble classifier based on paired sampling strategy for steganalysis. Experimental results showed that different sampling strategies could influence the performance of detection in steganalysis to some extent. Compared with the existing methods in steganalysis, the proposed method can significantly reduce the er- ror rate in detection performance of steganalysis system.
出处
《光电子.激光》
EI
CAS
CSCD
北大核心
2015年第4期746-751,共6页
Journal of Optoelectronics·Laser
基金
国家自然科学基金(61379152
61403417
61402530)
陕西省自然科学基金(2014JQ8301)资助项目
关键词
信息隐藏
隐写分析
成对采样
选择性集成
information hiding
steganalysis
paired sampling
selective ensemble