干涉式光学综合孔径压缩编码成像的方法研究

Optical synthesis aperture (OSA) interference imaging can be widely used in remote sensing, military reconnaissance and astronomy observation. Image reconstruction is one of the key issues. However, conventional reconstruction methods based on Nyquist sampling theory require large sample points and thus the reconstruction speed is slow, especially for large telescope interferometer in which data processing, storage and transmission is costly. For this, a novel optical synthesis aperture interference imaging method based on compressive coded aperture is presented, which can obtain higher image resolution from smaller sample points within shorter time, and can be easily used in practical applications. Three key issues are discussed in detail. Firstly, the principle of interference imaging via compressive coded aperture is demonstrated and the associated simulation experiments are presented. The impact of measurement matrix, aberrations, noise on reconstruction performance is also described. Secondly, the relationship between the array configuration and compressed sensing (CS) reconstruction algorithm is analyzed and a novel optimization criterion of array configuration for synthesis aperture based on CS theory is presented. Thirdly, the implemention of optical synthesis aperture interference imaging via compressive coded aperture in practical applications is discussed and a simple simulation system is built to verify the efficiency of the above research. Based on these work, a new type of optical synthesis aperture interference imaging with low-cost, high-speed and high-resolution characteristics in the future can be expected.

光学综合孔径干涉成像在遥感、军事侦察、天文观测等领域具有广泛的应用基础。图像重构是光学综合孔径干涉成像的关键问题。传统的基于奈奎斯特采样定理的重构算法所需采样点多，成像速度慢，特别是对于大型光干涉阵，数据处理、存储和传输的代价很大。本项目提出一种基于压缩编码的的光学综合孔径干涉成像方法，可以利用较少的采样点在较快的时间内获取较高的分辨率，且易于硬件实现。本项目将开展基于压缩编码理论的光学综合孔径干涉成像重构算法的理论论证与仿真分析，研究观测矩阵、系统误差以及噪声对重构算法的影响；分析综合孔径阵列结构和压缩感知重构性能的关系，提出新的基于压缩感知理论的阵列优化设计方法；探索压缩感知框架下光学综合孔径干涉成像的硬件实现方案，通过模拟验证装置，开展图像重构的实验研究。通过本项目关键问题的研究，将为未来实现低成本、高速、高分辨率的新型光综合孔径干涉成像系统奠定坚实的基础。

本项目提出基于压缩编码的光学综合孔径干涉成像方法，该方法可以利用较少的采样点在较快的时间内获取较高的分辨率，降低数据处理、存储和传输成本，适用于遥感、军事侦察等领域。我们按原定计划开展了研究，并基本取得了预期的研究成果：深入研究了制约压缩感知实际应用的测量矩阵设计方法，提出基于Tent、Colpitts、Boolen混沌序列的测量矩阵构造方法；提出基于混沌相位调制的光学压缩成像方法，该方法无需处理高维数据和向量间的相乘，节省计算时间和内存消耗，当探测器阵元为原来的1/4时，即可获得与原始图像视觉相当的重构图像；提出基于混沌相位调制的压缩编码光学综合孔径成像方法，通过建立数学模型，分析观测矩阵、系统误差以及噪声对重构性能的影响，得出该方法对系统误差和噪声具有很强的鲁棒性；提出OSA压缩编码成像的硬件模拟装置，但由于器件尚未全部到货，未展开实验研究。.