This paper considers the feature space of DT-MRI as a differential manifold with an affine-invariant metric. We generalise Di Zenzo’s structure tensor to tensor-valued images for edge detection. To improve the quality of the edges, we develop a generalised Perona-Malik method for smoothing tensor images. We demonstrate our algorithm on both synthetic and real DT-MRI data.


Feature Space Edge Detection Structure Tensor Multispectral Image IEEE Signal Processing Letter 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Baser, P.J., Mattiello, J., LeBihan, D.: Mr. diffusion tensor spectroscopy and imaging. Biophysical Journal 66, 259–267 (1994)CrossRefGoogle Scholar
  2. 2.
    O’Donnell, L., Grimson, W., Westin, C.-F.: Interface detection in diffusion tensor mri. In: Barillot, C., Haynor, D.R., Hellier, P. (eds.) MICCAI 2004. LNCS, vol. 3216, pp. 360–367. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  3. 3.
    Feddern, C., Weickert, J., Burgeth, B.: Level-set methods for tensor-valued images. In: Proc. IEEE Workshop on Variational, Geometric and Level Set Methods in Computer Vision, pp. 65–72Google Scholar
  4. 4.
    Di Zenzo, S.: A note on the gradient of a multi-image. Computer Vision, Graphics, and Image Processing 33, 116–125 (1986)CrossRefGoogle Scholar
  5. 5.
    Perona, P., Malik, J.: Scale-space and edge dectection using anisotropic diffusion. IEEE Trans. Pattern Anal. and Machine Intell. 12(7), 629–639 (1990)CrossRefGoogle Scholar
  6. 6.
    Fletcher, P., Joshi, S.: Principal geodesic analysis on symmetric spaces: Statistics of diffusion tensors. In: Proc. Computer Vision Approaches to Medical Image Analysis, ECCV Workshop (2004)Google Scholar
  7. 7.
    Lenglet, C., Rousson, M., Deriche, R., Faugeras, O.: A riemannian approach to diffusion tensor images segmentation. In: Christensen, G.E., Sonka, M. (eds.) IPMI 2005. LNCS, vol. 3565, pp. 591–602. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  8. 8.
    Pennec, X., Fillard, P., Ayache, N.: A riemannian framework for tensor computing. International Journal of Computer Vision 66(1), 41–66 (2006)CrossRefMathSciNetGoogle Scholar
  9. 9.
    Moakher, M.: A differential geometric approach to the geometric mean of symmetric positive-definite matrices. SIAM Journal on Matrix Analysis and Applications 26(3), 735–747 (2005)MATHCrossRefMathSciNetGoogle Scholar
  10. 10.
    Cumani, A.: Edge detection in multispectral images. CVGIP: Graphical Models and Image Processing 53(1), 40–51 (1991)MATHCrossRefGoogle Scholar
  11. 11.
    Sochen, N., Kimmel, R., Malladi, R.: A general framework for low level vision. IEEE Trans. on Image Processing 7(3), 310–318 (1998)MATHCrossRefMathSciNetGoogle Scholar
  12. 12.
    Tschumperle, D.: PDE’s Based Regularization of Multivalued Images and Applications. PHD Thesis, University of Nice-Sophia Antipolis (2002)Google Scholar
  13. 13.
    Sapiro, G., Ringach, D.L.: Anisotropic diffusion of multivalued images with application to color filtering. IEEE Trans. on Image Processing 5(11), 1582–1586 (1996)CrossRefGoogle Scholar
  14. 14.
    Blomgren, P., Chan, T.F.: Color tv: Total variation methods for restoration of vector-valued images. IEEE Trans. on Image Processing 7(3), 304–309 (1998)CrossRefGoogle Scholar
  15. 15.
    Tang, B., Sapiro, G., Caselles, V.: Diffusion of general data on non-flat manifolds via harmonic maps theory: The direction diffusion case. International Journal of Computer Vision 36(2), 149–161 (2000)CrossRefGoogle Scholar
  16. 16.
    Pardo, A., Sapiro, G.: Vector probability diffusion. IEEE Signal Processing Letters 8(4)Google Scholar
  17. 17.
    Catte, F., Lions, P.L., Morel, J.M., Coll, T.: Image selective smoothing and edge detection by nonlinear diffusion. SIAM J. Num. Anal. 29, 182–193 (1992)MATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Fan Zhang
    • 1
  • Edwin R. Hancock
    • 1
  1. 1.Department of Computer ScienceUniversity of YorkYorkUK

Personalised recommendations