High-dynamic range compressive spectral imaging by adaptive filtering

Compressive hyperspectral imaging systems (CSI) capture the three-dimensional (3D) information of a scene by measuring two dimensional (2D) using a small set of coded focal plane array (FPA) compressive measurement. A reconstruction algorithm takes advantage of the compressive measurements sparsity to recover the 3D data cube. Traditionally, CASSI uses block-unblock coded apertures to spatially modulate the light, the modulation has binary entries. In CASSI the quality of the reconstructed images depends on the design of these coded apertures and the FPA saturation. This work presents a new CASSI architecture based on grayscaled coded apertures (GCA) which reduce the saturation and increase the dynamic range of the FPA detector. The set of codes is calculated in a realtime adaptive manner such that the FPA compressive measurements are used to determine the structure of the GCA. Simulations show the improvement in the quality of the reconstructed images of the architecture based on GCA.