The cookie-related information is fully under our control. These cookies are not used for any purpose other than those described here. Unibo policy
Discrete tomography (DT) aims at retrieving object composed of materials with known constant properties from tomographic data. In general, the problem is combinatorial in nature and hence, NP-hard. In this poster, we are proposing the dual of such a problem with zero duality gap. The resulting problem is an L1-regularized least-squares problem that can be easily solved. Preliminary experiments show that the proposed approach can reconstructs the object almost perfectly from noisy projection data.
This is poster number 25 in Poster Session