1-1-10
5 CONCLUSION
In this paper, we presented a practical method for the
registration of 2.5D laser range image of urban objects.
The registration is performed at two levels: pair-wise
registration and multiple views’ registration. Pair-wise
registration is conducted using Z-lmages, where four
transformation parameters, a horizontal rotation angle and
three translation parameters, are tracked. Multiple views’
registration aims at solving the error accumulation
problem. It is solved as a least square minimization
problem. Through outdoor experiments, it is demonstrated
that the method has robustness and accuracy in the
sense of automation. The research also contributes to a
matching method using information theory. Although it
was originally motivated for the matching of Z-lmages, the
method is of general utility. It can be extended to the
problem domain of matching other kinds of data sets,
such as curved line or planar face characterized images
and so on.
Future research has to be addressed on the improvement
of accuracy in tracking translation parameter along Z-axis,
and minimizing the estimation error in global matching by
discriminate the contributions from different views of laser
range image.
6 REFERENCE
[1] B.Kamgar-Parsi, J.L.Jones, A.Rosenfeld, Registration
of multiple overlapping range image without
Distinctive Features, IEEE Trans.PAMI, vol.13, no.9,
1991.
[2] C.Thorpe, M.H.Hebert, T.Kanade, S.A.Shafer, “Vision
and Navigation for the Carnegie-Mellon Navlab”,
IEEE PAMI, vol.10, no.3, pp.363-373,1988.
[3] F.M.Ingels, “Information and Coding Theory”, Intext
Educational Publishers, Scranton, Pennsylvania,
1971.
[4] G.Champleboux, S.Lavallee, R.Szeliski, L.Brunie,
“From accurate range imaging sensor calibration to
accurate model-based 3-D object localization”, in
proceedings CVPR-92.
[5] G.Vosselman, Relational Matching, Lecture Notes in
Computer Science 628, Springer-Verlag, Berlin
Heidelberg, 1992.
[6] H.Shum, K.lkeuchi, R.Reddy, “Principle Component
Analysis with Missing Data and Its Application to
Object Modeling”, Proc.lEEE CVPR, 1994.
[7] H.Zhao, R.Shibasaki, Automated Registration of
Ground-Based Laser Range Image for
Reconstructing Urban 3D Object, IAPRS, Vol.32,
Part3-4W2, pp.27-34,1997.
[8] K.L.Boyer and A.C.Kak, Structural Stereopsis for 3-D
Vision, IEEE TPAMI, vol.10, no.2, pp.144-166, 1988.
[9] K.Kanatani, Geometric Computation for Machine
Vision, Oxford University Press, Oxford, U.K., 1993.
[10] K.Ng, V.Sequeira, S.Butterfield, D.Hogg,
J.G.M.Goncalves, An Integrated Multi-sensory
System for Photo-realistic 3D Scene Reconstruction,
IAPRS, vol.XXXII, part 5, pp356-363,1998.
[11] M.Okutomi, T.Kanade, A Multiple-Baseline Stereo,
IEEE Trans. PAMI, vol.15, no.4,1993.
[12] R.E.BIahut, “Principles and Practice of Information
Theory”, Addison-Wesley, Reading, MA, 1987.
[13] Sanfeliu and K.S.Fu, A Distance Measure between
Attributed Relational Graphs for Pattern Recognition,
IEEETSMC, vol.13, no.3, pp.353-362,1983.
[14] S.Ozawa, M.Notomi, H.Zen, A Wide Scope
Modeling to Reconstruct Urban Scene, IAPRS,
vol.XXXII, part 5, pp370-376,1998.
[15] W.Forstner, Image Analysis Techniques for Digital
Photogrammetry, Proceedings of the 42th
Photogrammetric Week, Series of the Institute of
Photogrammetry, Stuttgart Univ., vol.13, pp.205-221,
1989.
[16] Y.Chen, G.Medioni, “Object modelling by registration
of multiple range images”, image and vision
computing, vol.10, no.3, pp. 145-155,1992.