The registration of scattered point clouds with Gauss-Markoff model

For large or complex objects, they have to be scanned many times to recover its entire 3D shape in reverse engineering applications. The object surface can be sampled point by point using a structure light fringe projection. The method of least squares can match overlapping surfaces to estimate rotation matrix and translating vector between a local coordinate system and the template coordinate system. The Gauss-Markoff model can minimize the sum of squares of Euclidean distances between surfaces for matching 3D surface patches. This research uses the least squares method to the Gauss-Markoff model for the registration of point clouds. A vase registration example shows the feasibility of the proposed method. It takes about 4 seconds for the registration of 1531209 points with 0.027mm registration error. The surface profile after the registration is complete and smooth, which can meet the requirement of subsequent surface reconstruction.