编码孔径中红外光谱成像系统的光谱重构算法

针对一款基于双DMD的光谱维编码中波红外光谱成像系统的应用需求，在经典的Hadamard变换完备编码理论基础上，提出了一种具有最大信噪比增益的编码矩阵。结合基于双DMD的光谱维编码中波红外光谱成像系统的光路结构和工作原理，探讨了光谱维DMD所加载的编码模板的设计方法。建立了区块化光谱重构算法的数学模型，并进行了实验验证。实验结果表明，原理样机在具有良好的光谱成像能力的基础上，能够实现较高实时性的凝视成像，其在低光谱分辨率模式下光谱数据立方体帧频能够达到22.35帧/s,验证了区块化光谱重构算法的有效性。

Spectral Reconstruction Algorithm of Coded Aperture MWIR Spectral Imaging System

Aiming at the application requirements of a dual-DMD-based coded aperture mid-wave infrared spectral (MWIR) imaging system encoding in spectral dimension, an encoding matrix with maximum signal-to-noise ratio gain is proposed based on the classical Hadamard transform complete encoding theory. Combined with the optical path structure and working principle of the dual-DMD-based coded aperture MWIR spectral imaging system encoding in spectral dimension, the design method of the encoding mask loaded by the DMD which encodes in spectral dimension is discussed. The mathematical model of the block spectral reconstruction algorithm is established and verified by experiments. The experimental results show that the prototype can perform staring imaging with high real-time performance, under the premise of good spectral imaging capability. In the low spectral resolution mode, the frame rate of the spectral data cube can reach 22.35frames/s. Thus, the proposed block spectral reconstruction algorithm is verified to be effective and valuable.