为了解决编码感知快照光谱成像技术(CASSI)现有的光利用率低、重构精度低等问题，通过对CASSI系统光路的研究，提出了一种新型的高光利用率高精度的计算光谱成像技术；相比于CASSI系统增加了全色分支光路的采样，全色分支的采样与CASSI分支的采样在数据上形成互补。原理样机的成像实验表明，提出的系统大大减轻了重构问题的欠定程度；通过全色分支和CASSI分支的双路数据配准和融合，实现了高精度的光谱数据重构。相比于CASSI系统，所提出的系统光通量利用率提高了一倍以上。

In order to solve the issue of low utilization of light and low-precision reconstruction in the existing coded aperture snapshot spectral imager (CASSI), the computational spectrometry characteristic of high utilization of light and precision is developed through the research on CASSI. Compared with CASSI, sample of the full-color branch is added, which is complementary to the CASSI branch in data. Imaging experiment results show the underdetermined degree of reconstruction decreases sharply. By the data alignment and fusion between the full-color branch and CASSI branch, the high-precision data reconstruction is finished. As compared with CASSI, utilization of the optical throughput in the developed system is doubled.