Approximate Image Reconstruction in Landscape Reflection Imaging

Simple reflection imaging of landscape (scenery or extended objects) poses the inverse problem ofreconstructing the landscape reflectivity function from its integrals onsome particular family of spheres. Such data acquisition is encoded in the framework ofa Radon transform on this family of spheres. In spite of the existence of an exact inversionformula, the numerical landscape reflectivity function reconstitution is best obtainedwith an approximate but judiciously chosen reconstruction kernel. We describe theworking of this reflection imaging modality and its theoretical handling, introducean efficient and stable image reconstruction algorithm, and present simulationresults to prove the validity of this choice as well as to demonstrate the feasibility of thisimaging process.