Improving Registration of 3-D Medical Images Using a Mechanical Based Method

The registration of 3-D objects is an important problem in the medical imaging field. The problem arises with two images of the same modality taken from different positions or taken at different times (before and after an operation for example) or with two images of different modalities. We present a new method for computing the rigid transformation between two objects: we study the motion of a solid in a potential field instead of minimizing explicitly an energy. This article addresses a theoretical improvement of the method described in [1], some additional features like the undersampling of data or the multi-potentials approach. We present experimental results with real 3-D medical images.