Advertisement

Neural Networks for Image Restoration from the Magnitude of Its Fourier Transform

  • Adrian Burian
  • Jukka Saarinen
  • Pauli Kuosmanen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2085)

Abstract

In this paper the problem of image restoration from its Fourier spectrum magnitude is shown to be NP-complete. We propose the use of recurrent neural networks for solving the problem. The neural network incorporates the constants related to the real and imaginary parts of the image spectrum. The solution is provided by the steady state of the neural network, then is verified and eventually improved with the iterative Fourier transform algorithm. The obtained simulation results demonstrate the high efficiency of the proposed approach.

Keywords

Neural Network Recurrent Neural Network Implementation Purpose Maximum Entropy Method Nonlinear Optimization Problem 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bajkova, A.T.: Phase Retrieval Algorithm based on Maximum Entropy Method. In: Proc. of Int. Symp. on Signals, Systems and Electronics. (1995) 307–309Google Scholar
  2. 2.
    Burian, A., Kuosmanen, P., Rusu, C.: 1-D Non-Minimum Phase Retrieval by Gain Differences. In: Proc. of IEEE Int. Conf. on Electronics, Circuits and Systems, Vol. 3. Pafos, Cyprus, (1999) 1001–1005Google Scholar
  3. 3.
    Burian, A., Kuosmanen, P., Saarinen, J., Rusu, C.: Two Dimensional Phase Retrieval using Neural Networks. In: IEEE International Workshop on Neural Networks for Signal Processing. Sydney, December 2000, 652–662Google Scholar
  4. 4.
    Cantakeris, N.: Magnitude-only reconstruction of 2-D sequences with finite regions of support. In: IEEE Trans. Acoust., Speech and Signal Processing, Vol. 31. (1983) 1256–1262Google Scholar
  5. 5.
    Cichocki, A., Unbehauen, R.: Neural Networks for Optimization and Signal Processing. John Wiley & Sons, (1993)Google Scholar
  6. 6.
    Fienup, J.R.: Phase retrieval algorithms: a comparison. In: Applied Optics, Vol. 21, No. 15. (1982) 2758–2769Google Scholar
  7. 7.
    Garey, M.R., Johnson, D.S.: Computers and Intractability. A guide to the theory of NP-completeness. Bell Telephone Laboratories, Inc., (1979)Google Scholar
  8. 8.
    Mou-yan, Z., Unbehauen, R., Methods for Reconstruction of 2-D Sequences from Fourier Transform Magnitude. In: IEEE Trans. Image Processing, Vol. 6, No. 2. (1997) 222–233Google Scholar
  9. 9.
    Sahinoglou, Z., Cabrera, S.D., On Phase Retrieval of Finite-Length Sequences Using the Initial Time Sample. In: IEEE Trans. Circuits and Systems, Vol. 38, No. 8. August (1991) 954–958Google Scholar
  10. 10.
    Sauz J.L., Huang, T.S., Wu, T.F.: A note on iterative Fourier transform phase reconstruction from magnitude. In: IEEE Trans. on Acoustic, Speech, and Signal Processing, Vol. 32, No. 6. (1984) 1251–1254Google Scholar
  11. 11.
    Seldin, J.H., Fienup, J.R.: Numerical investigation of the uniqueness of phase retrieval. In: Journal Opt. Soc. Amer., Vol. 7, No. 1. (1990) 3–13Google Scholar
  12. 12.
    Wooshik, K.: Two-Dimensional Phase Retrieval using a Window Function. In: Proc. of IEEE Int. Conf. on Acoust., Speech and Signal Processing, Vol. 3. (1999) 1625–1628Google Scholar
  13. 13.
    Yagle A.E., Bell A.E.: One-and Two-Dimensional Minimum and Nonminimum Phase Retrieval by Solving Linear Systems of Equations. In: IEEE Trans. on Signal Processing, Vol. 47, No. 11. (1999) 2978–2989Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Adrian Burian
    • 1
  • Jukka Saarinen
    • 1
  • Pauli Kuosmanen
    • 1
  1. 1.Tampere University of TechnologyTampereFinland

Personalised recommendations