TWC: Medium: Steganography in Empirical Sources Using Acquisition Oracle

TWC：中：使用 Acquisition Oracle 在经验来源中进行隐写术

• 批准号: 1561446
• 负责人: Jessica Fridrich
• 金额: $ 57.59万
• 依托单位: SUNY at Binghamton
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-05-15 至 2021-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1561446&HistoricalAwards=false
• 关键词: TWC; Medium; Steganography; Empirical; Sources

The project focuses on advancing the field of digital image steganography -- a covert way of communication in which information is hidden in other objects, such as digital media files, to assure privacy. For a secure steganographic system, it should be impossible to prove the presence of hidden data. Achieving this level of security in practice is extraordinarily difficult because digital media is hard to describe using statistical models with accuracy necessary to guarantee perfect security. This project works with devices that acquire an image to learn a sufficiently accurate model within which it becomes possible to construct steganography with a verifiable level of security and performance that can be contrasted with theoretically achievable limits. The result is a novel advanced privacy tool needed in countries that prohibit the use of encryption or in a hostile environment when there is need to communicate without attracting attention via channels in control of an adversary or via public, insecure channels. On the other hand, a deeper understanding of the limits of covert communication will facilitate better defense against such methods of deception, an effort recognized as steganalysis. The project explores commercial applications of data hiding include signal authentication, integrity verification, and secure data dissemination. This research contributes to trusted information exchange, data mining, information assurance, network and computer security, counter-deception, and intrusion detection and its prevention.

该项目专注于推进数字图像隐写领域--一种隐蔽的通信方式，信息隐藏在其他对象中，如数字媒体文件，以确保隐私。对于一个安全的隐写系统，应该不可能证明隐藏数据的存在。在实践中实现这种级别的安全是非常困难的，因为数字媒体很难使用确保完美安全所需的准确统计模型来描述。该项目与获取图像的设备合作，以学习足够准确的模型，在该模型中，可以构建具有可验证的安全和性能级别的隐写术，该级别的安全性和性能可以与理论上可达到的极限进行对比。其结果是，在禁止使用加密的国家或在敌对环境中，当需要通过控制对手的渠道或通过公共、不安全的渠道进行通信而不引起注意时，需要一种新的高级隐私工具。另一方面，更深入地了解秘密通信的局限性将有助于更好地防御这种欺骗方法，这一努力被认为是隐写分析。该项目探索数据隐藏的商业应用，包括信号验证、完整性验证和安全数据传播。本研究在可信信息交换、数据挖掘、信息保障、网络与计算机安全、反欺骗、入侵检测及其预防等方面做出了贡献。