Iterative closest geometric objects registration

Abstract In this paper, closed-form solutions are obtained for registering two sets of line segments, triangle patches, or even general simple geometric objects that are defined by a set of ordered points. Based on these new registration approaches, the iterative closest line segment registration (ICL) algorithm and the iterative closest triangle patch registration (ICT) algorithm are developed similar to the ICP algorithm. To simplify the mathematical representation, the concept of matrix scalar product is defined and some of its properties are given. The newly developed registration methods are tested. The test shows that the ICL algorithm and the ICT algorithm work much better than the conventional ICP algorithm considering that the ICL and the ICT algorithms are much less sensitive to the initial orientations of the object.

[1]  K. S. Arun,et al.  Least-Squares Fitting of Two 3-D Point Sets , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Behrooz Kamgar-Parsi,et al.  Matching Sets of 3D Line Segments with Application to Polygonal Arc Matching , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Lisa M. Brown,et al.  A survey of image registration techniques , 1992, CSUR.

[5]  Berthold K. P. Horn,et al.  Closed-form solution of absolute orientation using unit quaternions , 1987 .

[6]  Max A. Viergever,et al.  A survey of medical image registration , 1998, Medical Image Anal..

[7]  M. Hebert,et al.  The Representation, Recognition, and Locating of 3-D Objects , 1986 .

[8]  S. Umeyama,et al.  Least-Squares Estimation of Transformation Parameters Between Two Point Patterns , 1991, IEEE Trans. Pattern Anal. Mach. Intell..