项目名称: 非整数预测误差扩展无损信息隐藏关键技术研究
项目编号: No.61272414
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 自动化技术、计算机技术
项目作者: 项世军
作者单位: 暨南大学
项目金额: 80万元
中文摘要: 无损信息隐藏技术可用于无损压缩、完整性认证和容错等目的,是信息隐藏研究中的一个重要分支。当前的预测误差扩展嵌入策略存在一定问题,且缺少用于认证和容错的无损信息隐藏技术。本项目针对信息隐藏的应用前景,重点研究大容量、高保真的无损信息隐藏算法和容错技术。主要内容包括:1)研究基于非整数预测误差扩展的无损嵌入策略,解决当前预测误差扩展算法只能对整数误差进行扩展的问题;2)以音频/图像为载体,研究全方向上下文预测器,解决当前预测误差扩展算法中预测器只能使用前面已知样本进行预测的问题;3)结合密码哈希、无损压缩和纠错码等技术,研究新的无损嵌入策略,设计可用于完整性认证和具有一定失真修复能力的容错方案(如音频在一定篡改失真下的无损恢复)。通过研究,试图解决无损信息隐藏在嵌入量和容错性方面的几个关键问题,发展和完善基于预测误差扩展的无损信息隐藏理论,为数字媒体的信息隐藏容错技术做出理论上和方法上的贡献。
中文关键词: 信息隐藏;可逆水印;非整数预测;同态加密;加密域
英文摘要: In the information hiding community, lossless data hiding (also called reversible watermarking) for digital media is an important research issue. It allows to embed useful information in a host signal without any loss of host information. In practice, this technique could be applied for improving lossless compression, fault-tolerant ability and authentication of meida integrity. In the literature, prediction-error based reversible watermarking algorithms can not expand non-integer prediction errors for data embedding. Also, there are lack of high-capacity and lossless data hiding algorithms to enhance the fault-tolerant ability of digital audio for cropping and replacement attacks. Towards these practical requirements on information hiding techniques, this research project aims at developing high-capacity and high-fidelity lossless data hiding algorithms, and lossless data hiding based content integrity verification and error-tolerance techniques. This research project is composed of three main components: 1) To develop new lossless data hiding algorithms based on non-integer prediction error expansion, to solve the problem that the existing approaches can not expand non-integer prediction values; 2) To design new prediction models to solve the problem that in the existing prediction-error expansion embedding ap
英文关键词: Information Hiding;Reversible Watermarking;Non-integer Prediction;Homomorphic Encryption;Encrypted Domain