Knowledge and Gestalt Interaction

  • Eckart MichaelsenEmail author
  • Jochen Meidow
Part of the Advances in Computer Vision and Pattern Recognition book series (ACVPR)


This chapter first introduces knowledge utilization on pictorial data. That is, it introduces inference. It turns out that conventional deductive inference is in this case of little use. Instead abductive inference is used, which brings with it certain risks of failure. The Gestalt-laws proved quite universal and stable inferences from parts to aggregates. They can thus be included in knowledge-based image analysis systems, as kind of default robust constructions. Then only the domain specific knowledge parts must be added. Two examples for such cooperation between perceptual grouping along the laws of Gestalt operations on one side and automatic knowledge utilization on the other hand are given, both on remotely sensed data: 1) Thermal hyper-spectra are analyzed. These are given by an aerial spectrometer on the geographic plane. On this plane Gestalt organization can to a certain degree recognize certain repetitive patterns in a hierarchy, while knowledge about urban objects and their mutual organization, as well as knowledge about spectra of certain materials, can be utilized for classification. 2) The synthetic aperture radar data used as example for lattice grouping


  1. 1.
    Schenk T (1995) A layered abduction model of building recognition. In: Automatic extraction of man-made objects from aerial and space images, Ascona workshop of the ETH Zurich, pp 117–123CrossRefGoogle Scholar
  2. 2.
    Niemann H. (1990) Pattern analysis and understanding. SpringerGoogle Scholar
  3. 3.
    Matsuyama T, Hwang VS-S (1990) SIGMA, a knowledge-based aerial image understanding system. SpringerGoogle Scholar
  4. 4.
    Hinz S, Baumgartner A, Steger C, Mayer H, Eckstein W, Ebner H, Radig B (1999) Road extraction in rural and urban areas. In: Förstner W, Liedtke C-E, Bückner J (eds) Semantic modelling for the acquisition of topographic information from images and maps (SMATI 1999), pp 133–153Google Scholar
  5. 5.
    Sörgel U (ed) (1990) Radar remote sensing of urban areas. SpringerGoogle Scholar
  6. 6.
    Pizlo Z, Li Y, Sawada T, Steinman RM (2014) Making a machine that sees like us. Oxford University PressGoogle Scholar
  7. 7.
    Leyton M (2014) Symmetry, causality, mind. MIT Press, Cambrige, MaGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Fraunhofer IOSBEttlingenGermany

Personalised recommendations