基于有效载体标注和融合态载体构造的低密度信息隐藏方法研究

In order to meet the demand for decentralized and low-density information hiding from larger confidential data transmission, we try to propose some multi-carrier-based information hiding algorithms rather than the single-carrier-based ones. First, aiming at identification of some effective carriers in certain kinds of carrier groups which consist of many digital images and 3D mesh models, we need to analyze some entropy and geometric attributes. Second, we should extract some components from these effective carriers as the input to build the fusion-state carrier based on some arithmetic and logic operation. Last, the embedding data will be written in the fusion-state carrier according to the consistency of entropy and geometry. The search can be summarized as follows. (1) The labelling of the effective carriers and the extracting of the similarities between them for the fusing of multi-carrier. (2) The construction of the fusion-state carrier. (3) Low-density data embedding based on fusion-state carrier. During the research, we will put forward some effective carrier identification methods using similarity parameters, and build multi-carrier fusion operation models. Besides, we will present the difference region division of fusion-state carrier and the demand-based data embedding schemes according to the consistency of entropy and geometry. The research results can provide valuable reference models for the low-density carrier-based information hiding, and will be helpful to the national security.

课题针对基于单载体的信息隐藏难以满足较大数据量秘密信息的分散式、低密度隐藏需求的现状，以数字图像和三维网格模型的熵值和几何属性为突破口、以寻找有效载体间的相似性指标为手段，按熵值比例从有效载体中提取部分信息作为构造融合态载体的输入量，最后围绕载、密一致性原则进行低密度信息隐藏。课题的研究内容概括为：面向多载体融合的有效载体标注及载体间相似性指标提取、多载体融合构造和秘密信息的低密度写入等三点。课题将提出基于相似性的有效载体标注方法并构建同位/异位、同轴/非同轴的多载体融合运算器模型。在此基础上，借助熵值和几何属性一致性原则，提出多载体融合态差异性分区和信息按需写入算法。研究成果可为基于载体的低密度信息隐藏提供有价值的参考模型，且对国家安全有积极的意义。