Fuzzy techniques for image enhancement and reconstruction

A fuzzy-based enhancement algorithm that considers multiple-object images is presented. This algorithm is based on crisping the boundaries of the objects by minimizing their compactness. This concept is incorporated in an iterative reconstruction-reprojection algorithm which achieves a compromise between the boundary crispness of the reconstructed image and the fidelity of the available projections in incomplete angular coverage situations. Simulation results are provided to demonstrate the effectiveness of the algorithm.<>