| Summary: | This paper presents a fast algorithm for the registration of shapes implicitly represented by their characteristic functions. The proposed algorithm aims to recover the transformation parameters (scaling, rotation, and translation) by minimizing a dissimilarity term between two shapes. The algorithm is based on phase correlation and statistical shape moments to compute the registration parameters individually. The algorithm proposed here is applied to various registration problems, to address issues such as the registration of shapes with various topologies, and registration of complex shapes containing various numbers of sub-shapes. Our method proposed here is characterized with a better performance for registration over large databases of shapes, a better accuracy, a higher convergence speed and robustness at the presence of excessive noise in comparison with other state-of-the-art shape registration algorithms in the literature
|
|---|
