Hostname: page-component-586b7cd67f-rdxmf Total loading time: 0 Render date: 2024-11-20T06:40:39.361Z Has data issue: false hasContentIssue false

Super Resolution by Phase-Gradient Unravelling

Published online by Cambridge University Press:  19 July 2016

D. Fraser
Affiliation:
Department of Electrical Engineering, University of New South Wales, ADFA, Canberra, ACT, Australia
B. R. Hunt
Affiliation:
Department of Electrical and Computer Engineering, University of Arizona, Tucson, Arizona, U.S.A.

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

In this work, we present another proof of the theoretical existence of super resolution, under certain conditions. The proof relies on an ideal model in which a prototypical discrete image is formed by summation of many discrete pulses placed anywhere on a regular grid. If the model is then band-limited in spatial frequency, the original, grid-resolution pulse image may be reconstructed from the band-limited information. The reconstruction uses “phase-gradient unravelling” — i.e., from a very limited number of terms of the discrete Fourier transform of an image, which defines a very limited spatial-frequency band, we extract or unravel the individual phase-gradients which, together, define the original image.

Type
Imaging Theory
Copyright
Copyright © Kluwer 1994 

References

[1] Frieden, B.R., and Zolatin, C.K.: 1986, “Monte Carlo restoration of binary objects”, J. Opt. Soc. Am. A, 3, pp. 731734.CrossRefGoogle Scholar
[2] Gerchberg, R.W.: 1989, “Superresolution through error function extrapolation”, IEEE Trans. on ASSP, 37, pp. 16031606.Google Scholar