Abstract
The classical principle axes theory (PAT) is generalized to affine transformations with a new parametrization approach. Based on this extended theory, a fast multi-scale technique is derived for the alignment of affine objects which is at least by one magnitude more accurate than the results of the classical PAT [1]–[8]. Compared to the algorithms of Cygansky & Orr [7] and Faber & Stokeley [8] the technique is not sensitive to noise [9] or to symmetries of the objects, since the transformation is derived from a second order moment tensor. In addition, it is shown with the extension of the theory, that the application of the PAT described in [1] by Bajcsi & Kovacic results in strong rotational and scaling misalignment (with rotational errors up to 45°), which can be completely suppressed by the generalized theory based on an appropriate parametrization and optimization of a similarity criterion.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
R. Bajcsy, S. Kovacic,”Multiresolution elastic matching,” Comput Vision Graph Im Proc, vol. 46, 1–21, 1989
L.S. Hibbard, R.A. Hawkins, ”Objective image alignment for 3-D reconstruction of digital autoradiograms”, J Neurosci Methods, vol. 26, 55–74, 1988
A.W. Toga, P.K Banerjee, ”Registration Revisited”, J Neurosci Methods, vol. 48, 1–13, 1993
E.J. Holupka, H.M. Kooy, ”A geometric algorithm for medical image correlations,” Med Phys, vol. 19, 433–438, 1992
N.M. Alpert, J.F. Bradshaw, ”The principle axes transformation: a method for image registration,” J Nucl Med, vol. 31, 1717–1722, 1990
A.P. Dhawan, L.K. Arata, A.V. Levy, J. Manta, ”Iterative principle axes registration method for analysis of MR-PET brain images,” IEEE Trans Biomed Eng, vol. 45, 1079–1087, 1995
D. Cygansky, J.A. Orr, ”Application of tensor theory to object recognition and orientation determination”, IEEE Trans Pattern Anal Mach Intell, vol. 7, 662–673, 1985
T.L. Faber, E.M. Stokeley, ”Orientation of 3-D structures in medical images”, IEEE Trans Pattern Anal Mach Intell, vol. 20, 626–633, 1988
Y.S. Abu-Mostafa, D. Psaltis, ”Image normalization by complex moments”, IEEE Trans Pattern Anal Mach Intell, vol. 7, 46–53, 1985
T. Schormann, A. Dabringhaus, K. Zilles, ”Statistics of deformations in histology and Improved Alignment with MRI,” IEEE Transactions on Medical Imaging, vol. 14, 25–35, 1995
T. Schormann, M. v. Matthey, A. Dabringhaus, K. Zilles K. Zilles K, ”Alignment of 3-D brain data sets originating from MR and histology,” Bioimaging, vol. 1, 119–128, 1993; ERRATUM, Bioimaging, vol. 1, 185, 1993
T. Schormann, S. Henn, K. Zilles, ”A new approach to fast elastic alignment with applications to human brains,” Lecture Notes in Computer Science, vol. 1131, 337–342, 1996
S. Geyer, A. Ledberg, A. Schleicher, S. Kinomura, T. Schormann, U. Biirgel, T. Klingberg, J. Larsson, K. Zilles, P.E. Roland, ”Two different areas within the primary cortex of man,” Nature, vol. 382, 805–807, 1996
G. Fischer, Analytische Geometrie II, Braunschweig: Vieweg & Sohn, 1979
B. Jâhne, Digitale Bildverarbeitung, Berlin: Springer-Verlag, 1989
W.H. Press, B.P. Flannery, S.A. Teukolsky, W.T. Vetterling, Numerical recipes in C, Cambridge: Cambridge U. P., 1988
A. Budó: Theoretische Mechanik, Berlin: Deutscher Verlag der Wissenschaften, 1980
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Schormann, T., Dabringhaus, A., Zilles, K. (1997). Extension of the Principle Axes Theory for the Determination of Affine Transformations. In: Paulus, E., Wahl, F.M. (eds) Mustererkennung 1997. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-60893-3_41
Download citation
DOI: https://doi.org/10.1007/978-3-642-60893-3_41
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63426-3
Online ISBN: 978-3-642-60893-3
eBook Packages: Springer Book Archive