双极性压缩观测光谱成像技术研究

摘要

There is an increasing optical imaging revolution based on compressed sensing(CS). CS is a mathematical framework with several powerful theorems that provide insight into how a high resolution image can be inferred better than large-format focal plane arrays(FPAs). Practical architectures which have been developed to exploit CS theory have to face the challenges of photon efficiency, noise, non-negative restriction. A physically realizable system was proposed in this paper based on compressive sensing optical imaging scheme. The method utilizes two-way optical architecture to implement bipolar observation and resolve inconsistencies between the theoretical requirements and physical constraints in compressing imaging(CI). Multiplexing and multivalve mark strategy makes wide view imaging and multivalve observation possible. As enough measurements can be obtained in a single exposure, such system can reconstruct original image precisely without additional information. Simulated experiments demonstrate effectiveness and robustness of the proposed method.%压缩传感突破了 Nyquist-Shannon 采样定理的限制，从随机观测的少量测量值中即可高精度地获取图像，给成像设备的设计带来了巨大变革。压缩传感理论证明可以通过重构恢复获得比焦平面阵列分辨率更高的场景图像。尽管在理论上存在巨大优越性，压缩传感成像系统的物理实现需要考虑一些实际问题。文章围绕压缩传感成像设备进行了研究，提出了一种物理可实现的压缩成像方法，该方法利用双通道观测架构实现压缩成像中的双极性观测，解决了压缩成像理论与实际物理约束不一致的问题。采用多路技术和多膜技术实现大视场观测与多次视场观测，该方法可以单次曝光获取充足的观测值来高精度重构原图像。压缩光谱成像数值仿真实验验证了该方法的有效性和鲁棒性。