Elementary Morphology for SO(2)- and SO(3)-Orientation Fields

In this article, techniques to process fields of special orthogonal matrices by means of mathematical morphology are proposed. Since the group structure of SO(2)- resp. SO(3)-fields is not suitable to establish useful notions of infimum and supremum, they are transformed into scalar resp. Sym(2)-fields utilizing a matrix-valued version of the Cayley transform. However, for symmetric 2×2 matrices, that is for Sym(2)-fields, elementary morphological operations are available, see [6] and [7]. Several examples and numerical results are presented to show the merits and the limitations of this novel approach. Additionally, the series of transformations utilized in the proposed methods open up possibilities to visualize SO(2)- and SO(3)-fields.

[1]  Remco Duits,et al.  Sharpening Fibers in Diffusion Weighted MRI via Erosion , 2014, Visualization and Processing of Tensors and Higher Order Descriptors for Multi-Valued Data.

[2]  Jos B. T. M. Roerdink,et al.  Group-Invariant Frames for Colour Morphology , 2013, ISMM.

[3]  Joachim Weickert,et al.  Morphology for Higher-Dimensional Tensor Data Via Loewner Ordering , 2005, ISMM.

[4]  V. Barnett.  The Ordering of Multivariate Data , 1976 .

[5]  Krishnamoorthy Sivakumar,et al.  Morphological Operators for Image Sequences , 1995, Comput. Vis. Image Underst..

[6]  Pierre Soille,et al.  Morphological Image Analysis , 1999 .

[7]  Joachim Weickert,et al.  Mathematical morphology for tensor data induced by the Loewner orderingin higher dimensions , 2005 .

[8]  G. Matheron.  Éléments pour une théorie des milieux poreux , 1967 .

[9]  Edward J. Delp,et al.  Morphological operations for color image processing , 1999, J. Electronic Imaging.

[10]  Preprint Nr,et al.  Mathematical Morphology for Tensor Data Induced by the Loewner Ordering in Higher Dimensions , 2005 .

[11]  Thomas Brox,et al.  A TV Flow Based Local Scale Measure for Texture Discrimination , 2004, ECCV.

[12]  Sébastien Lefèvre,et al.  A comparative study on multivariate mathematical morphology , 2007, Pattern Recognit..

[13]  H. Heijmans.  Morphological image operators , 1994 .

[14]  Bernhard Burgeth,et al.  An approach to color-morphology based on Einstein addition and Loewner order , 2014, Pattern Recognit. Lett..