A Feature Based Method for Rigid Registration of Anatomical Surfaces

The problem is: given two three-dimensional (3D) triangulated surfaces A, B find the best rigid transformation that brings surface A as close as possible to surface B. The most commonly used method in the literature is iterative closest point (ICP). ICP uses either the whole surface description or feature points on the surfaces. The drawback of ICP is that it requires a good initial estimation to achieve convergence. This paper proposes a variation of ICP, randomized iterative closest curve (RICC). The input of RICC is feature curves of the input surfaces. The input surfaces, in our examples, are 3D triangulated brain structures. The feature curves are mathematically based shape features, crest lines that have anatomical significance on brain structures. The advantage of RICC versus ICP is that it does not need an initial estimation to converge to a global minimum. Also, RICC exploits the structurally meaningful feature lines instead of using just the points that are structurally meaningless. Randomization is used to boost its performance in hard cases. Experimental results show that RICC achieves fast convergence in all cases, regardless of input surfaces’ pose.

[1]  M. Loew,et al.  Automated registration of brain images using edge and surface features. , 1999, IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society.

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

[3]  Ioannis Pitas,et al.  Morphological Iterative Closest Point Algorithm , 1997, CAIP.

[4]  Yi Ping Hung,et al.  A fast automatic method for registration of partially-overlapping range images , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[5]  W. EggertD.,et al.  Estimating 3-D rigid body transformations , 1997 .

[6]  Georgios Stylianou,et al.  Shape Feature Extraction , 2003 .

[7]  C. Q. Little,et al.  Registration of range data using a hybrid simulated annealing and iterative closest point algorithm , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[8]  Jérôme Declerck,et al.  Automatic registration and alignment on a template of cardiac stress and rest reoriented SPECT images , 1997, IEEE Transactions on Medical Imaging.

[9]  Patrick J. Flynn,et al.  Range image registration: A software platform and empirical evaluation , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

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

[11]  Georgios Stylianou.  Automatic crest line extraction from anatomical surfaces , 2003, SPIE Medical Imaging.

[12]  Nicholas Ayache,et al.  Rigid, affine and locally affine registration of free-form surfaces , 1996, International Journal of Computer Vision.

[13]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[14]  Benoit M. Dawant,et al.  Registration of 3-D images using weighted geometrical features , 1996, IEEE Trans. Medical Imaging.

[15]  Zhengyou Zhang,et al.  Iterative point matching for registration of free-form curves and surfaces , 1994, International Journal of Computer Vision.

[16]  Robert B. Fisher,et al.  Estimating 3-D rigid body transformations: a comparison of four major algorithms , 1997, Machine Vision and Applications.

[17]  Richard A. Volz,et al.  Estimating 3-D location parameters using dual number quaternions , 1991, CVGIP Image Underst..

[18]  Leif Kobbelt,et al.  Extraction of feature lines on triangulated surfaces using morphological operators , 2000 .

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

[20]  Sang Wook Lee,et al.  ICP Registration Using Invariant Features , 2002, IEEE Trans. Pattern Anal. Mach. Intell..