Machine Vision and Applications

, Volume 8, Issue 5, pp 262–274 | Cite as

Feature extraction and image segmentation using self-organizing networks

  • Yong -Jian Zheng


Feature extraction and image segmentation (FEIS) are two primary goals of almost all image-understanding systems. They are also the issues at which we look in this paper. We think of FEIS as a multilevel process of grouping and describing at each level. We emphasize the importance of grouping during this process because we believe that many features and events in real images are only perceived by combining weak evidence of several organized pixels or other low-level features. To realize FEIS based on this formulation, we must deal with such problems as how to discover grouping rules, how to develop grouping systems to integrate grouping rules, how to embed grouping processes into FEIS systems, and how to evaluate the quality of extracted features at various levels. We use self-organizing networks to develop grouping systems that take the organization of human visual perception into consideration. We demonstrate our approach by solving two concrete problems: extracting linear features in digital images and partitioning color images into regions. We present the results of experiments on real images.

Key words

Feature extraction Image segmentation Grouping Perceptual organization Self-organizing networks 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Boldt M, Weiss R, Riseman E (1989) Token-based extraction of straight lines. IEEE Trans Syst Man Cyber 19:1581–1694Google Scholar
  2. 2.
    Burns JB, Hanson AR, Riseman EM, (1986) Extracting straight lines. IEEE Patt Anal Machine Intell 8:425–455Google Scholar
  3. 3.
    Carpenter GA, Grossberg S (1988). The ART of adaptive pattern recognition by a self-organizing neural network. Computer 21:77–88Google Scholar
  4. 4.
    Duda RO, Hart PE (1972) Use of the Hough transformation to detect lines and curves in pictures. Commun ACM 15:11–15Google Scholar
  5. 5.
    Grossberg S (1976) Adaptive pattern classification and universal recoding, I: parallel development and coding of neural feature detectors. Biol Cybern 23:121–134Google Scholar
  6. 6.
    Grossberg S (1987) Competitive learning: from interactive activation to adaptive resonance. Cogn Sci 2:23–63Google Scholar
  7. 7.
    Grossberg S, Mingolla E (1985) Neural dynamics of perceptual grouping: textures, boundaries, and emergent segmentations. Perception Psychophysics 38:141–171Google Scholar
  8. 8.
    Hanson A, Riseman E (1987) The VISIONS image understanding system. In: Brown C (ed) Advances in computer vision. Erlbaum Press, Hillsdalle, N.J.Google Scholar
  9. 9.
    Herman M, Kanade T (1986) Incremental reconstruction of 3D scenes from multiple, complex images. Artif Intell 30:289–341Google Scholar
  10. 10.
    Huertas A, Nevatia R (1988) Detecting buildings in aerial images. Computer Vision Graph Image Processing 41:131–152Google Scholar
  11. 11.
    Huber PJ, (1981) Robust statistics. Wiley, New YorkGoogle Scholar
  12. 12.
    Khan GM, Gillies DF (1992) Extracting contours by perceptual grouping. Image Vision Comput 10:77–88Google Scholar
  13. 13.
    Lowe DG (1985) Perceptual organization and visual recognition. Kluwer Academic, Hingham, MAGoogle Scholar
  14. 14.
    Lu HQ, Aggarwal KK (1992) Applying perceptual organization to the detection of man-made objects in non-urban scenes. Patt Recogn 25:835–853Google Scholar
  15. 15.
    Marr D, Hildreth E (1980) Theory of edge detection. Proc Roy Soc Lond B 207:187–217Google Scholar
  16. 16.
    Marr D (1982) Vision. Freeman, San Francisco.Google Scholar
  17. 17.
    Medioni G, Nevatia R (1984) Matching images using linear features. IEEE Patt Anal Machine Intell 6:675–685Google Scholar
  18. 18.
    Minsky ML (1975) A framework for representing knowledge. In: Winston PH (ed) The psychology of computer vision. McGraw-Hill, New YorkGoogle Scholar
  19. 19.
    Mohan R, Nevatia R (1989) Using perceptual organization to extract 3-D structures. IEEE Part Anal Machine Intell 11:1121–1139Google Scholar
  20. 20.
    Nevatia R, Babu KR (1980) Linear feature extraction and description. Comput Graph Image Processing 13:257–269Google Scholar
  21. 21.
    Ohta Y (1985) Knowledge-based interpretation of outdoor natural color scenes. Pitman Advanced Publishing, Boston MassGoogle Scholar
  22. 22.
    Philippe PO, Dubes RC (1992) Performance evaluation for four classes of textural features. Patt Recogn 25:819–833Google Scholar
  23. 23.
    Riseman EM, AR Hanson (1988) A methodology for the development of general knowledge-based vision systems. In Arbib MA (ed.) Vision systems and cooperative computation, MIT Press, Combridge Mass, pp 285–313Google Scholar
  24. 24.
    Sarkar S, Boyer KL (1993) Integration, inference, and management of spatial information using bayesian network: perceptual organization. IEEE Trans Patt Anal Machine Intell 15:256–274Google Scholar
  25. 25.
    Venkates war V, Chellappa R (1990) Extraction of straight lines in aerial images. Procedings of the 5th European Signal Processing Conference, Barcelona, Spain, pp 1671–1674Google Scholar
  26. 26.
    Wertheimer, M (1923) Laws of organization in perceptual forms. Psychologische Forschung 4:301–350Google Scholar
  27. 27.
    Wilson R, Span M (1988) Image segmentation and uncertainty. Research Studies Press, Wiley, ChichesterGoogle Scholar
  28. 28.
    Zeidenberg M (1990) Neural network models in artificial intelligence. Ellis Horwood, West Sussex, EnglandGoogle Scholar
  29. 29.
    Zheng Y.-J, Hahn M. (1990) Surface reconstruction from images in the presence of discontinuities, occlusions and deformations. Int Arch Photogrammetry Remote Sensing 28:1121–1144Google Scholar
  30. 30.
    Zheng Y.-J (1992) Inductive inference and inverse problems in computer vision. In: Förstner/Ruwiedel (eds) Robust computer vision, Wichmann, KarlsruheGoogle Scholar

Copyright information

© Springer-Verlag 1995

Authors and Affiliations

  • Yong -Jian Zheng
    • 1
  1. 1.Daimler-Benz AGResearch Center UlmUlmGermany

Personalised recommendations