Fuzzy Vector Objective Optimization Algorithm for Image Reconstruction from Incomplete Projections

This paper deals with the problem of image reconstruction from incomplete projections. A novel fuzzy vector objective optimization model is developed by integrating the fuzzy set theory and vector objective optimization (multi-objective decision-making). The objective function is expressed as a membership function, and the minimum operator is taken as a fuzzy operator. Furthermore, a novel iterative method is proposed to resolve the fuzzy optimization problem. The images reconstructed from simulated noise projections and real projections obtained from an industrial scanner show that the new algorithm can provide higher resolution and better smoothness than the images reconstructed by the transformation method and the conventional iterative method, so it is more feasible for image reconstruction from incomplete projections.