Compressive spectral video by dynamic spatial-spectral-temporal windowed codification

Snapshot compressive imaging of high-dimensional data is indispensable for numerous areas of study. Despite the remarkable advances in either spectral or temporal approaches separately, optically compressing a spectral video in a single snapshot remains challenging. In this article, we report a novel compressive spectral video scheme that offers a dynamic color-coded wavelength coding plus a windowing approach to handle the temporal dimension. The proposed compressive coding scheme is implemented by synchronizing a tunable bandpass filter and a coded aperture device to code and acquire the spectral video within a fixed exposure time. Special attention is given to the design and evaluation of the dynamic coded apertures using binary masks. We develop and implement a plug-and-play alternating-direction method of multiplier to efficiently recover the four-dimensional datacubes. We demonstrate the ability of the proposed approach to capture and reconstruct spectral video information in both spectral classification and motion tracking applications.