Iconic registration based on the local dissimilarity card

In this work, we focus on iconic registration as one of the fundamental tasks in medical imaging. We have proposed a new approach that consists of the registration of two mono-modal images based on the local dissimilarity card (LDC). Instead of maximizing the similarity between the two input images as in traditional registration techniques, the main contribution of this work is minimizing the dissimilarity between them. Experiments in RMI medical images have shown that the proposed approach is as reliable as an original method using mutual information-based registration as a reference.