Fast and robust registration of 3D surfaces using low curvature patches

The paper describes a novel range data registration algorithm, specifically designed for accuracy, speed, and robustness. Like many recent registration techniques, our Robust-Closest-Patch algorithm (RCP) iteratively matches model patches to data surfaces based on the current pose and then re-estimates pose based on these matches. RCP has several novel features: 1) online registration is driven by low curvature patches computed from the model offline; 2) an approximate normal distance between a patch and a surface is used, avoiding the need to estimate local surface normal and curvature from noisy data; 3) pose is solved exactly by a linear system in six parameters, using a symmetric formulation of the rotation constraint; 4) robustness is ensured using an M-estimator that estimates both the rigid pose parameters and the error standard deviation. Results are shown using models and range data from turbine blade inspection.

[1]  C. Jennison,et al.  Robust Statistics: The Approach Based on Influence Functions , 1987 .

[2]  Tomas Lozano-Perez,et al.  An automatic registration method for frameless stereotaxy, image guided surgery, and enhanced reality visualization , 1996 .

[3]  J. Tukey,et al.  The Fitting of Power Series, Meaning Polynomials, Illustrated on Band-Spectroscopic Data , 1974 .

[4]  Evangelos E. Milios,et al.  Robot Pose Estimation in Unknown Environments by Matching 2D Range Scans , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[5]  P. Holland,et al.  Robust regression using iteratively reweighted least-squares , 1977 .

[6]  Gérard G. Medioni,et al.  Object modelling by registration of multiple range images , 1992, Image Vis. Comput..

[7]  Werner A. Stahel,et al.  Robust Statistics: The Approach Based on Influence Functions , 1987 .

[8]  Naokazu Yokoya,et al.  A Robust Method for Registration and Segmentation of Multiple Range Images , 1995, Comput. Vis. Image Underst..

[9]  Robert Bergevin,et al.  Towards a General Multi-View Registration Technique , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

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

[11]  Chitra Dorai,et al.  Optimal registration of multiple range views , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[12]  Katsushi Ikeuchi,et al.  Building 3-D Models from Unregistered Range Images , 1995, CVGIP Graph. Model. Image Process..

[13]  P. Rousseeuw.  Least Median of Squares Regression , 1984 .

[14]  Hong-Tzong Yau,et al.  Automated precision measurement of surface profile in CAD-directed inspection , 1992, IEEE Trans. Robotics Autom..

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

[16]  Gérard G. Medioni,et al.  Object modeling by registration of multiple range images , 1991, Proceedings. 1991 IEEE International Conference on Robotics and Automation.

[17]  Allen R. Hanson,et al.  Robust methods for estimating pose and a sensitivity analysis , 1994 .

[18]  J. Stuelpnagel.  On the Parametrization of the Three-Dimensional Rotation Group , 1964 .

[19]  I. Faux,et al.  Computational Geometry for Design and Manufacture , 1979 .

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

[21]  Katsushi Ikeuchi,et al.  Building 3-D models from unregistered range images , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.