负载自适应的批量隐写研究

为了进一步提高批量隐写的安全性, 针对以往自适应批量隐写方案的不足设计了一种简单可靠的新方案。首先借助基于随机森林的集成分类器确定当前隐写分析技术下图像的隐写容量, 在此基础上确定各个图像嵌入的信息量, 方案最大程度地利用了载体资源, 并通过对秘密信息进行分割分组进一步增强了安全性。实验结果表明, 隐写容量计算中的安全判定方法准确性高, 在保证低漏检率的同时避免了较高的虚警率, 且运行时间更短。

Research on payload adaptive batch steganography

In order to improve the security of batch steganography, this paper proposed a novel adaptive batch steganography system. First, it estimated embedding capacity of images under current steganalysis techniques by bringing ensemble classifier based on random forest into steganalysis, then distributed message bits to the images according to their embedding capacities. The new system could make full use of the images and improve batch steganography security further by dividing the message into groups. Experimental results show that the new security definition decreases the total detection error and running time, havi-ng a good balance between missed detection rate and false alarm rate.