Iterative closest point

The Iterative Closest Point Algorithm is an algorithm which allows to adapt point clouds together.

For the point cloud coordinate transformations are determined such that the distances are minimized between the point clouds. When it is determined for each point from a point cloud of each of the next point ( closest point) of the other point cloud. The sum of the squares of the distances is minimized by adjustment of the transformation parameters. This operation is done iteratively until the optimum is found.

The algorithm is mainly used for the registration of 3D laser scans. The 3D point clouds from different viewpoints can be adjusted by the ICP together. Thus, a complete model is composed of individual scans. Another field of application is the localization in robotics, a sub-problem of Simultaneous Localization and Mapping.

420879
de