基于两步广义相移干涉术和压缩鬼成像的光学信息安全技术的研究

As the optical information security has the obvious advantages such as the ability of high-speed parallel processing and high degree of freedom for encryption, the studies of data encryption, authentication and hiding based on optical information processing techniques have been particularly demonstrated to be an important field extensively developed. Based on two-step generalized phase-shifting interferometry and compressive ghost imaging, the studies of this project include following aspects: (1) we will deign and optimize the algorithms of blind phase-shift extraction and wavefront reconstruction in the two-step generalized phase-shifting interferometry, combined with compressive sensing theory, we will propose the optical cryptosystem and authentication with high security, we will also propose the watermarking scheme based on singular value decomposition. (2) We will propose the optical cryptosystem based on compressive ghost imaging and Logistic map chaos algorithm, the multiple-image encryption-decryption method based on position or wavelength multiplexing techniques will also be proposed. (3) We will deign and optimize the orthogonal matching pursuit algorithm in compressive sensing, combined with Lifting Wavelet transform and XOR operation, we will propose the optical cryptosystem method with low transmission load based on computational ghost imaging. (4) We will propose hierarchical multiple-image authentication method based on random amplitude sampling and phase information multiplexing, we will design hyperplane secret sharing algorithm, we will also propose the novel hierarchical authentication scheme which can solve the problem of multiple keys management and hierarchical access control. We believe that the research achievements can promote the application of optical information security.

由于光学信息安全系统具有高并行度和高加密维度等优点，因此，探索基于光学信息处理技术的数据加密、认证及隐藏方案成为活跃开展的一个研究领域。本项目拟围绕两步广义相移干涉术和压缩鬼成像，开展如下几个方面的研究：（1）设计和优化两步广义相移盲取及波前再现算法，结合压缩感知理论，开发安全强度更高的光学密码和认证系统，提出基于奇异值矩阵分解的数字水印方案；（2）提出基于压缩鬼成像和Logistic映射混沌算法的光学密码方案，结合位置复用和波长复用技术，提出多幅图像的加解密方法；（3）设计和优化压缩感知理论中的正交匹配追踪算法，结合提升小波变换与异或操作，提出低传输负荷的压缩鬼成像光学密码方法；（4）结合振幅稀疏数据采样和相位信息复用等技术，提出新型的多幅图像分级认证方法，设计超平面秘密共享算法，提出可以解决多重密钥管理和分级访问控制的分级认证新方案。本项目的研究成果将会大大推进光学信息安全技术的应用。

由于光学信息安全系统具有高并行度和高加密维度等优点，因此，探索基于光学信息处理技术的数据加密、认证及隐藏方案成为活跃开展的一个研究领域。本项目在前期的工作基础之上，主要围绕压缩两步相位数字全息干涉术和单像素成像理论和应用研究，取得了一系列的研究成果，主要包括：(1)基于压缩两步相移数字全息干涉术，提出了两级图像认证方法和平滑压缩信号空间复用的多幅图像加密方法；(2)基于奇异值分解鬼成像，提出了光学图像编码和加密方法、光学图像信息隐藏系统和(t, n)门限秘密共享算法的图像分级认证方法；(3)基于压缩单像素成像理论，提出了稀疏特性的多幅图像编码与加密方法和超平面密钥分享的光学多级认证方法。本项目很好地实现了预定目标，取得的研究成果对发展新型光学信息安全的理论和应用具有重要的实际意义。