| | This Article | |
| |
| |
| | Share | |
| |
| |
| | Bibliographic References | |
| |
| |
| | Add to: | |
| |
 Digg
 Furl
 Spurl
 Blink
 Simpy
 Google
 Del.icio.us
 Y!MyWeb
| |
| | Search | |
| |
| |
| | |
Image Representation by Sign Information
July 1989 (vol. 11 no. 7)
pp. 729-738
Image representation by sign data in its most general context is considered for the case when f is a two-dimensional signal. The authors study conditions under which f is determined by its sign or, in other words, by its quantization to 1 bit of information. This study is carried out in two different directions. First, theoretical results are presented that set an algebraic condition under which real zero-crossings uniquely specify a band-limited image. An interesting paradigm arising in theoretical computer vision is then posed: are the zero-crossing of f convolved with a Laplacian-of-a-Gaussian at a single channel enough for unambiguously representing f? Second, the problem of the completeness of the representation when the position of the zero-crossings is known only approximately is addressed. It is shown that when sign(f) is sampled, significant ambiguities are introduced in the representation. Experimental results are presented which were obtained from an iterative algorithm devised to reconstruct real images from multiscale sign information.
[1] 729B. Logan, "Information in the zero crossings of bandpass signals,"Bell Syst. Tech. J., vol. 56, no. 4, pp. 487-510, Apr. 1977.[2] N. Levinson, "Density and gap theorems,"Amer. Math. Soc. Collq. Pub., XXVI, New York, pp. 25-26, 1940.[3] D. Marr, S. Ullman, and T. Poggio, "Bandpass channels, zero crossings, and early visual information processing,"J. Opt. Soc. Amer., vol. 69, pp. 914-16, June 1979.[4] D. Marr and E. Hildreth, "Theory of edge detection,"Proc. Rev. Soc. Lond., vol. B 207, pp. 187-217, 1980.[5] D. Marr,Vision. San Francisco, CA: Freeman, 1982.[6] R. Young, "The Gaussian derivative model for machine vision: Visual cortex simulation,"Res. Publ., Comput. Sci. Dep., General Motors, GMR-5323, July 1986.[7] A. Yuille and T. Poggio, "Fingerprint theorems for zero crossings,"J. Opt. Soc. Amer., vol. 2, no. 5, pp. 683-692, May 1985.[8] S. Curtis, A. Oppenheim, and J. Lim, "Signal reconstruction from Fourier transform sign information,"IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, pp. 643-657, June 1985.[9] S. Shitz and Y. Zeevi, "On the duality of time and frequency domain signal reconstruction from partial information,"IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, pp. 1486-1498, Dec. 1985.[10] H. Voelcker and A. Requicha, "Clipping and signal determination: Two algorithms requiring validation,"IEEE Trans. Commun., vol. COM-21, pp. 738-744, June 1973.[11] B. Logan, "Signals designed for recovery after clipping--I. Localization of infinite products,"AT&T Bell Lab. Tech. J., vol. 63, pp. 261-285, Feb. 1984.[12] B. Logan, "Signals designed for recovery after clipping--II. Fourier transform theory of recovery,"AT&T Bell Lab. Tech. J., pp. 287-306, Feb. 1984.[13] D. Rotem, and Y. Y. Zeevi, "Image reconstruction from zero-crossings,"IEEE Trans. Acoustics Speech Signal Processing, vol. ASSP-34, no. 5, pp. 1269-1277, Oct. 1986.[14] A. Zakhor and D. Izraelevitz, "A note on the sampling of zero crossings of two-dimensional signals,"Proc. IEEE, vol. 74, pp. 1285- 1287, Sept. 1986.[15] T. T. Huang, J. Sanz, and W. E. Blanz, "A study on image reconstruction from one-bit Fourier phase information: Theory, sampling considerations, and coherent image model," submitted toIEEE Trans. ASSP; also presented at the Int. Conf. ASSP, Dallas TX, Apr. 1987.[16] R. Hummel, "Representations based on zero crossings in scale-space," inIEEE Comput. Soc. Conf. Comput. Vision Pattern Recognition, 1986, pp. 204-209.[17] R. Haralick, "Digital step edges from zero crossings of second directional derivatives,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-6, pp. 58-68, Jan. 1984.[18] M. Shah, A. Sood, and R. Jain, "Pulse and staircase models for detecting edges at multiple resolution," Cent. Res. Integr. Manufacturing, Univ. Michigan, Tech. Rep. RSD-TR-7-85, June 1985.[19] A. Yuille and T. Poggio, "Scaling theorems for zero crossings,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-8, no. 1, pp. 15-25, 1986.[20] A. Witkin, "Scale-space filtering," inProc. Int. Joint Conf. Artificial Intell., Karlsruhe, pp. 1019-1021, 1983.[21] J. Babaud, A. P. Witkin, M. Baudin, and R. O. Duda, "Uniqueness of the Gaussian kernel for scale-space filtering,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-8, pp. 26-33, Jan. 1986.[22] L. I. Ronkin,Introduction to the Theory of Entire Functions of Several Variables, Translations of Mathematical Monographs, Vol. 44, Amer. Math. Soc., Providence, RI, 1974.[23] W. F. Osgood,Funktionentheorie, Vol. 2. Chelsea, NY 1965.[24] R. Gerchberg and W. Saxton, "A practical algorithm for the determination of phase from image and diffraction plane pictures,"Optik, vol. 35, pp. 237-246, 1972.[25] A. Mostowski and M. Stark,Introduction to Higher Algebra. New York: Macmillan, 1964.[26] L. Hormander,Linear Partial Differential Operators. New York: Springer-Verlag, 1962.[27] J. L. C. Sanz, F. Cukierman, and T. S. Huang, "Stability of unique Fourier transform phase reconstruction,"J. Opt. Soc. Amer., vol. 73, no. 11, pp. 1442-1445, Nov. 1983.[28] J. L. C. Sanz and T. S. Huang, "Polynomial systems of equations and its applicability to the study of the effect of noise on multidimensional Fourier transform phase retrieval from magnitude,"IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, pp. 997- 1004, Aug. 1985.[29] B. Fuks,Introduction to the Theory of Analytic Functions in Several Complex Variables, Vol. 8, translations of theAmer. Math. Soc., Providence, RI, 1963.[30] J. L. C. Sanz and T. S. Huang, "Unique reconstruction of a bandlimited multidimensional signal from phase or magnitude,"J. Opt. Soc. Amer., vol. 73, no. 11, pp. 1446-1450, Nov. 1983.[31] W. Lawton and J. Morrison, "Factoring trigonometrical polynomials regarded as entire functions of exponential type,"J. Opt. Soc. Amer., to be published.[32] J. R. Fienup, "Phase retrieval: Algorithm improvements, uniqueness, and complex objects,"J. Opt. Soc. Amer., pp. 40-43.[33] R. Lane, W. Fright, and R. Bates, "Direct phase retrieval,"IEEE Trans. Acoust., Speech, Signal Processing, to be published.[34] D. Izraelevitz and J. S. Lim, "A new algorithm for closed form image reconstruction from Fourier transform magnitude," inDigest of Topical Meeting on Signal Recovery and Synthesis II. Washington, DC: Optical Society of America, 1986.[35] R. Nevatia, personal communication, USC, 1986.[36] J. Reimer and P. Lawrence, "CharacterizingΔGfiltered images by their zero crossings," presented atIEEE Int. Conf. ASSP, Dallas, TX, Apr. 1987.[37] F. Campbell and J. Robson, "Application of Fourier analysis to the visibility of gratings,"J. Phys.(London). vol. 197, pp. 551-556, 1968.[38] H. Wilson and J. Bergen, "A four mechanism model for spatial vision,"Vision Res., vol. 19, pp. 19-32, 1979.[39] J. L. C. Sanz, "Multidimensional signal representation by zero crossings: An algebraic study,"SIAM J. Appl. Math., to be published.[40] S. Curtis, Ph.D. dissertation, Dep. Elec. Comput. Eng., M.I.T., 1985.
Index Terms:
image reconstruction; image representation; sign information; two-dimensional signal; zero-crossings; band-limited image; computer vision; Laplacian-of-a-Gaussian; computer vision; iterative methods
Citation:
J.L.C. Sanz, T.T. Huang, "Image Representation by Sign Information," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, no. 7, pp. 729-738, July 1989, doi:10.1109/34.192467
