Advertisement

A statistical classification method for hierarchical irregular objects

  • Markus Peura
Poster Session B: Active Vision, Motion, Shape, Stereo
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1310)

Abstract

This paper introduces a method for classifying structured visual objects that appear frequently in meteorological, medical and biological imagery. The focused objects are taken to be highly irregular and composed of subobjects in a hierarchical manner. The approach consists three principal steps. At first, hierarchical objects are detected in an segmented image. Secondly, shape descriptors are used to extract information of the contours of the objects. Finally, a global description for an object is obtained by applying statistical moments. As the goal is to classify natural objects, the most challenging task is to tolerate irregularity present in both spatial and hierarchical levels. Experiments with artificial images show that the method combines succesfully shape descriptors and object hierarchy.

Keywords

Natural Object Shape Descriptor Tree Hierarchy Focus Object Applied Object 
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.

References

  1. 1.
    L. Gupta and M.D. Srinath. Contour sequence moments for the classification of closed planar shapes. Pattern, Recognition, 20(3):262–272, 1987.Google Scholar
  2. 2.
    Robert M. Haralick, K. Shamnugam, and Its'hak Dinstein. Textural features for image classification. IEEE Transactions on Systems, Man, and, Cybernetics, SMC-6(6):610–621, November 1973.Google Scholar
  3. 3.
    Chih-Cheng Heu, Wesley W. Chu, and Ricky K. Taira. A knowledge-based approach for retrieving images by content. IEEE Transactions on Knowledge and Data Engineering, 8(4):522–532, August 1996.Google Scholar
  4. 4.
    Ming-Kei Hit. Visual pattern recognition by moment invariants. IRE Transactions on Information Theory, 8(2):179–187, 1962.Google Scholar
  5. 5.
    Markus Peura and Jukka Iivarinen. Efficiency of simple shape descriptors. In IWVF3, 3th International Workshop on Visual Forms, Capri, Italy, May 1997.Google Scholar
  6. 6.
    Milan Sonka, Vaclav Hlavac, and Roger Boyle. Image Processing, Analysis and Machine Vision. Chapman-Hall Computing, London, 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Markus Peura
    • 1
  1. 1.Laboratory of Computer and Information ScienceHelsinki University of TechnologyHUTFinland

Personalised recommendations