Recalage hétérogène de nuages de points 3D.

The registration of two 3D point clouds is an essential step in many applications. The objective of our work is to estimate the isometric transformation to merge two heterogeneous point clouds obtained from two different sensors. In this paper, we present a new approach for 3D-3D registration which is distinguished by the nature of the extracted signature on each point and by the similarity criterion used to measure the degree of similarity. The descriptor that we propose is invariant to the rotation and also to the translation and overcomes the problem of multi-resolution that is related to heterogeneous data. At the end, our approach has been tested on synthetic data and applied on heterogeneous real data. [ABSTRACT FROM AUTHOR]