Spherical Shading Correction of Eye Fundus Image by Parabola Function

  • Kozo Okazaki
  • Shinichi Tamura


In this paper we are presenting four methods of correcting spherical shading for eye fundus images by using a parabola function. The estimation of the correcting function is difficult since the characteristics of the shading are unknown and we must estimate the parameters from a limited field photograph.
  1. (1)

    The hill climbing method; all the results obtained don’t converge into the true values, but they depend rather on the initial values.

  2. (2)

    The pattern search method; this is a simplification of the hill climbing method. The results obtained do not always converge into the true values.

  3. (3)

    The razor search method; to improve the convergence of the pattern search method, we apply the razor search method additionally to the pattern search method.

  4. (4)

    The slice method; when we binarize a spherical shaded image of eye fundus photograph by thresolding, a circular area is obtained. We can estimate the shading correcting function from this area. By this algorithm, we obtained the good corrected images that depict the blood vessels.



Pattern Search Circular Area Parabola Function Ocular Fundus Slice Method 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    B. H. McCormic, J. S. Read, R. T. Borovec, B. C. Amendola, and M. H. Goldbaum, “Image Processing in Television Ophthalmoscopy,” Digital Processing of Biomedical Images, K. Preston, Jr. and M. Onoe, eds., pp. 399–424, University of Tokyo Press, Tokyo, 1976.Google Scholar
  2. [2]
    S. Yamamoto and H. Yokouchi, “Automatic Recognition of Color Fundus Photographs,” Digital Processing of Biomedical Images, K. Preston, Jr. and M. Onoe, eds., pp. 385–398, University of Tokyo Press, Tokyo, 1976.Google Scholar
  3. [3]
    K. Akita and H. Kuga, “Towards understanding color ocular fundus images,” Proceedings of the International Joint Conference on Artificial Intelligence, Tokyo, pp. 7–12, 1979.Google Scholar
  4. [4]
    R. S. Ledley, “Computerized Pattern Recognition in Health Care Delivery,” Health Hand Book, G. K. Chacko, ed.,pp. 533–567, North-Holland, Amsterdam, 1979.Google Scholar
  5. [5]
    M. Zingirian, C. Braccini, S. Gaglio, G. Marino, and V. Tagliasco, “A processing system oriented to the management of images acquired by different ophthalmological diagnosis methods,” Proceedings of Computers in Ophthalmology, IEEE, pp. 111–122, 1979.Google Scholar
  6. [6]
    J. S. Read, A. C. Petersen, B. H. McCormic, and M. F. Goldberg, “The television ophthalmoscope image processor: Methods and applications,” Proceedings of Computers in Ophthalmology, IEEE, pp. 123–132, 1979.Google Scholar
  7. [7]
    M. Tanaka, S. Tamura, and K. Tanaka, “A technique for the automatic assembly of eye fundus photographs using blood vessel structure,” Proceedings of the International Conference on Cybernetics and Society, Tokyo and Kyoto, pp. 266–270, 1978.Google Scholar
  8. [8]
    S. Tamura, K. Tanaka, S. Ohmori, K. Okazaki, A. Okada, and M. Hoshi, “Semiautomatic leakage analyzing system for time series fluorescein ocular fundus angiography,” Pattern Recognition, 16 (2), pp. 149–162, 1983.CrossRefGoogle Scholar
  9. [9]
    R. Hooke and T. A. Joeves, “Direct search solution of numerical and statistical problems,” J. ACM, 8 (2), p. 212, 1961.MATHCrossRefGoogle Scholar
  10. [10]
    J. W. Bandler and P. A. MacDonald, “Optimization of microwave networks by razor search,” IEEE Trans. Microwave Theory & Tech., MTT-17(8), p. 552, 1969.Google Scholar

Copyright information

© Plenum Press, New York 1985

Authors and Affiliations

  • Kozo Okazaki
    • 1
  • Shinichi Tamura
    • 2
  1. 1.Department of Electrical Engineering Faculty of EngineeringTottori UniversityTottori, Tottori 680Japan
  2. 2.Department of Information and Computer Sciences Faculty of Engineering ScienceOsaka UniversityToyonaka, Osaka 560Japan

Personalised recommendations