Skip to Main Content
The basic problem of LINC-NIRVANA (LN) imaging is to combine the different images for getting a single image, possibly with a resolution close to that of a 22.8 m mirror in all directions. This result can be hardly reached in practice because it depends on the level and uniformity of the adaptive optics (AO) correction and on the declination of the scientific target, controlling the orientations of the baseline that can be used during its observation. In this article, we give a brief introduction to this problem. It can be solved by iterative methods related to the well-known Richardson-Lucy (RL) algorithm. Since RL is a particular case of the maximum likelihood method introduced by Shepp and Vardi in emission tomography and denoted expectation maximization (EM), improvements of EM, proposed in the field of medical imaging can be applied to LN.