Compressed sensing for earth observation with high resolution satellite imagery

In the field of satellite imagery, good image acquisition chains are constructed such that the produced image being transmitted from the satellite to the ground is well sampled (in order to achieve a posteriori good restoration results for example). This means that the Nyquist-Shannon sampling theorem is satisfied [1]. However this theorem does not take into account the properties of the signal to be transmitted. Some recent works have used prior knowledge of the signal properties such as its sparsity. This new theory has recently been extensively used in the literature since it offers nice mathematical results for acquiring and reconstructing a sparse or compressible signal. This method has already been used in remote sensing, especially in the Herschel mission, which is a satellite dedicated to observation of the universe. In this work, we propose to analyse the applicability of compressed sensing in the framework of earth observation with high resolution satellite imagery, considering simultaneously the impact on the image quality performances, the on-board capacities and the adaptability to on-board hard physical constraints.