ICV and fine-registration algorithms for an efficient merging of point clouds

One of most important and frequents needs that arise when we have to handle the point clouds is to merging them in an efficient and precise manner so that we can obtain more complex and bigger 3D models by chunks, seamlessly. ICV, Intelligent Cloud Viewer, is a software developed in-house by AESEI (spin-off of Polytechnic of Bari) that contains many tools and filters for point cloud processing. In this paper we illustrate the fine-registration algorithm implemented in ICV, one of the countless variants of ICP methods (Iterative Closets Points). The fine-registration method can be usefully adopted to estimate the differences between point clouds of the same subject executed in different times or for a comparison of point clouds of same subject obtained with different acquisition techniques, for example a laser-scanner and a photogrammetric technique. Furthermore, ICV integrates the M3C2 module for a robust error analysis, where by error is intended the differences or distances between similar point clouds.