On multiple spectral dependent blurring kernels for super-resolution and hyperspectral imaging

A new approach to perform the acquisition and the reconstruction of spatially super-resolved hyperspectral images is presented. The proposed hyperspectral sensing strategy is based on acquiring several low-resolved grayscale images following a specific acquisition scheme which takes profit from different spectral dependent blurring kernels. The proposed model describes how output grayscale pixels are related to input multi-spectral pixels. By its nature, an optical scattering media with particular spectral properties could possibly generate such appropriate Point Spread Functions (PSFs). This proposed technique claims that for well chosen scattering media, a common grayscale image sensor can be employed to acquire super-resolved hyperspectral images. The proposed reconstruction is performed by a regularization algorithm adding proper constraints on the hyperspectral cube. More generally, this can be considered as a hyperspectral compressive sensing problem since the number of measurements can be less than the total amount of reconstructed information.