Abstract
Fourier transform magnitudes are commonly used in the generation of templates in pattern recognition applications. We report on recent advances in Fourier phase retrieval which are relevant to pattern recognition. We emphasise in particular that the intrinsic form of a finite, positive image is, in general, uniquely related to the magnitude of its Fourier transform. We state conditions under which the Fourier phase can be reconstructed from samples of the Fourier magnitude, and describe a method of achieving this. Computational examples of restoration of Fourier phase (and hence, by Fourier transformation, the intrinsic form of the image) from samples of the Fourier magnitude are also presented.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Altmann J, Reitboeck HJ (1984) A fast correlation method for scale- and translation-invariant pattern recognition. IEEE Trans PAMI 6:46–57
Bates RHT (1978) On phase problems II. Optik 51:223–234
Bates RHT (1982) Astronomical speckle imaging. Phys Rep 90:203–297
Bates RHT (1984) Uniqueness of solutions to two-dimensional Fourier phase problem for localised and positive images. Comp Vision Graph Im Proc 25:205–217
Bates RHT, Fright WR (1983) Composite two-dimensional Fourier phase-restoration procedure. J Opt Soc Am 73:358–365
Bates RHT, Fright WR (1984) Reconstructing images from their Fourier intensities. In: Huang TS (ed) Advances in computer vision and image processing, vol 1: Image reconstruction from incomplete observations. JAI Press, Greenwich, Conn., pp 227–264
Bates RHT, Fright WR (1985) Two-dimensional phase restoration. In: Price JF (ed) Fourier techniques and applications. Plenum Press, New York, pp 121–148
Born M, Wolf E (1976) Principles of optics, 4th edn. Pergamon, London
Bracewell RN (1978) The Fourier transform and its applications, 2nd edn. McGraw-Hill, New York
Brigham EO (1974) The fast Fourier transform. Prentice Hall, Englewood Cliffs, NJ
Casasent D, Psaltis D (1976) Position, rotation, and scale invariant optical correlation. Appl Optics 15:1795–1799
Cavanagh P (1978) Size and position invariance in the visual system. Perception 7:167–177
DeVelis JB, Reynolds GO (1967) Theory and applications of holography. Addison-Wesley, Reading, Mass.
Fienup JR (1982) Phase retrieval algorithms: a comparison. Appl Optics 21:2758–2769
Fienup JR (1984) Experimental evidence of the uniqueness of phase retrieval from intensity data. In: Roberts JA (ed) Indirect imaging. Cambridge University Press, pp 99–109
Fright WR, Bates RHT (1982) Fourier phase problems are uniquely solvable in more than one dimension. III Computational examples for two dimensions. Optik 62:219–230
Gabor D (1972) Holography. Proc IEEE 60:655–668
Goodman JW (1968) Introduction to Fourier optics. McGraw-Hill, New York
Leith EN, Upatnieks J (1964) Wavefront reconstruction with diffuse illumination and three-dimensional objects. J Opt Soc Am 54:1295–1301
Menzel E, Mirande W, Weingartner I (1973) Fourier-Optik und Holographie. Springer, Wien New York
Reitboeck HJ, Altmann J (1984) A model for size- and rotation-invariant pattern processing in the visual system. Biol Cybern 51:113–121
Slepian D (1976) On bandwidth. Proc IEEE 64:292–300
Taylor LS (1981) The phase retrieval problem. IEEE Trans AP-29:386–391
Won MC, Mnyama D, Bates RHT (1985) Improving initial phase estimates for phase retrieval algorithms. Opt Acta 32:377–396
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Gardenier, P.H., McCallum, B.C. & Bates, R.H.T. Fourier transform magnitudes are unique pattern recognition templates. Biol. Cybern. 54, 385–391 (1986). https://doi.org/10.1007/BF00355544
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF00355544