Nonlinear Functionals in the Construction of Multiscale Affine Invariants

  • Esa Rahtu
  • Mikko Salo
  • Janne Heikkilä
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4522)


In this paper we introduce affine invariants based on a multiscale framework combined with nonlinear comparison operations. The resulting descriptors are histograms, which are computed from a set of comparison results using binary coding. The new constructions are analogous to other multiscale affine invariants, but the use of highly nonlinear operations yields clear advantages in discriminability. This is also demonstrated by the experiments, where comparable recognition rates are achieved with only a fraction of the computational load. The new methods are straightforward to implement and fast to evaluate from given image patches.


Local Binary Pattern Binary Code Computational Load Invariant Feature Invariant Moment 
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.
    Hu, M.: Visual pattern recognition by moment invariants. IEEE Trans. Information Theory 8, 179–187 (1962)Google Scholar
  2. 2.
    Flusser, J., Suk, T.: Pattern recognition by affine moment invariants. Pattern Recognition 26(1), 167–174 (1993)CrossRefMathSciNetGoogle Scholar
  3. 3.
    Ben-Arie, J., Wang, Z.: Pictorial recognition of objects employing affine invariance in the frequency domain. IEEE Trans. Pattern Analysis and Machine Intelligence 20(6), 604–618 (1998)CrossRefGoogle Scholar
  4. 4.
    Yang, Z., Cohen, F.S.: Cross-weighted moments and affine invariants for image registration and matching. IEEE Trans. Pattern Analysis and Machine Intelligence 21(8), 804–814 (1999)CrossRefGoogle Scholar
  5. 5.
    Petrou, M., Kadyrov, A.: Affine invariant features from the trace transform. IEEE Trans. Pattern Analysis and Machine Intelligence 26(1), 30–44 (2004)CrossRefGoogle Scholar
  6. 6.
    Rahtu, E., Salo, M., Heikkilä, J.: Affine invariant pattern recognition using multiscale autoconvolution. IEEE Transactions on Pattern Analysis and Machine Intelligence 27(6), 908–918 (2005)CrossRefGoogle Scholar
  7. 7.
    Rahtu, E., Salo, M., Heikkilä, J.: A New Efficient Method for Producing Global Affine Invariants. In: Roli, F., Vitulano, S. (eds.) ICIAP 2005. LNCS, vol. 3617, pp. 407–414. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  8. 8.
    Rahtu, E., Salo, M., Heikkilä, J.: Multiscale autoconvolution histograms for affine invariant pattern recognition. In: Proc. the 16th British Machine Vision Conference, vol. 3, Edinburgh, UK, pp. 1059–1068 (2006)Google Scholar
  9. 9.
    Ojala, T., Pietikäinen, M., Mäenpää, T.: Multiresolution gray-scale and rotation invariant texture classification with Local Binary Patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence 24(7), 971–987 (2002)CrossRefGoogle Scholar

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Esa Rahtu
    • 1
  • Mikko Salo
    • 2
  • Janne Heikkilä
    • 1
  1. 1.Machine Vision Group, Department of Electrical and Information Engineering, P.O. Box 4500, 90014 University of OuluFinland
  2. 2.Department of Mathematics and Statistics / RNI, P.O. Box 68, 00014 University of HelsinkiFinland

Personalised recommendations