Advertisement

Hybrid-Maximum Neural Network for Depth Analysis from Stereo-Image

  • Łukasz Laskowski
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6114)

Abstract

In present paper, we describe completely innovation architecture of artificial neural nets based on Hopfield structure for solving of stereo matching problem. Hybrid neural network consists of classical analogue Hopfield neural network and maximal neural network. The role of analogue Hopfield network is to find of attraction area of global minimum, whereas maximum network is to find accurate location of this minimum. Presented network characterizes by extremely high rate of working with the same accuracy as classical Hopfield-like network. It is very important as far as application and system of visually impaired people supporting are concerned. Considered network was taken under experimental tests with using real stereo pictures as well simulated stereo images. This allows on calculation of errors and direct comparison to classic analogue Hopfield neural network. Results of tests have shown, that the same accuracy of solution as for continuous Hopfield-like network, can be reached by described here structure in half number of classical Hopfield net iteration.

Keywords

Hopfield neural networks stereovision depth analysis hybrid network 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Aleksander, I.: Artificial vision for robots. Korgan Page (1983)Google Scholar
  2. 2.
    Barlow, H.B., Blackmore, C., Pettigrew, J.D.: The natural mechanism of binocular depth discrimination. J. Physiology 193, 327–342 (1967)Google Scholar
  3. 3.
    Faugeras, O.: Three-dimensional computer vision. A geometric viewpoint. MIT Press, Cambridge (1993)Google Scholar
  4. 4.
    Alvarez, L.: Dense Disparity Map Estimation Respecting Image Discontinuities: A PDE and Scale-Space Based Approach. Journal of Visual Communication and Image Representation 13, 3–21 (2002)CrossRefGoogle Scholar
  5. 5.
    Hopfield, J.J., Tank, D.W.: Neural computation of decisions in optimization problems. Biological Cybernetics 52, 141–152 (1985)zbMATHMathSciNetGoogle Scholar
  6. 6.
    Hopfield, J.J., Tank, D.W.: Artificial neural networks. IEEE Circuits and Devices Magazine 8, 3–10 (1988)CrossRefGoogle Scholar
  7. 7.
    Amit, D.J., Gutfreung, H., Sampolinsky, H.: Spin-glass models of neural networks. Physical Review A2 32(2), 1007–1018 (1985)CrossRefGoogle Scholar
  8. 8.
    Takefuji, Y., Lee, K.C., Aiso, H.: An artificial maximum neural network: a winner-take-all neuron model forcing the state of the system in a solution domain. Biological Cybernetics 67(3), 243–251 (1992)zbMATHCrossRefGoogle Scholar
  9. 9.
    Takefuji, Y., Lee, K.C.: Neural network computing for knight’s tour problems. Neurocomputing 4, 249–254 (1992)zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Łukasz Laskowski
    • 1
  1. 1.Department of Computer EngineeringTechnical University of CzestochowaCzestochowaPoland

Personalised recommendations