International Journal of Computer Vision

, Volume 34, Issue 2–3, pp 97–122 | Cite as

Are Edges Incomplete?

  • James H. Elder
Article

Abstract

We address the problem of computing a general-purpose early visual representation that satisfies two criteria. 1) Explicitness: To be more useful than the original pixel array, the representation must take a significant step toward making important image structure explicit. 2) Completeness: To support a diverse set of high-level tasks, the representation must not discard information of potential perceptual relevance. The most prevalent representation in image processing and computer vision that satisfies the completeness criterion is the wavelet code. In this paper, we propose a very different code which represents the location of each edge and the magnitude and blur scale of the underlying intensity change. By making edge structure explicit, we argue that this representation better satisfies the first criterion than do wavelet codes. To address the second criterion, we study the question of how much visual information is lost in the representation. We report a novel method for inverting the edge code to reconstruct a perceptually accurate estimate of the original image, and thus demonstrate that the proposed representation embodies virtually all of the perceptually relevant information contained in a natural image. This result bears on recent claims that edge representations do not contain all of the information needed for higher level tasks.

edge detection image reconstruction scale space diffusion blur deblurring denoising diffusion 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Adelson, E. 1991. The plenoptic function and the elements of early vision. In Computational Models of Visual Processing, M. Landy and J. Movshon (Eds.), MIT Press: Cambridge, MA.Google Scholar
  2. Adelson, E., Simoncelli, E., and Hingorani, R. 1987. Orthogonal pyramid transforms for image coding. In Proc. of SPIE, Vol. 845, Cambridge, MA, pp. 50–58.Google Scholar
  3. Adini, Y., Moses, Y., and Ullman, S. 1997. Face recognition: the problem of compensating for illumination changes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(7):721–732.Google Scholar
  4. Barlow, H. 1961. The coding of sensory messages. In Current Problems in Animal Behavior, W. Thorpe and O. Zangwill (Eds.), Cambridge U. Press, pp. 331–360.Google Scholar
  5. Barrow, H. and Tenenbaum, J. 1981. Computational vision. Proc. IEEE, 69:572–595.Google Scholar
  6. Barth, E., Caelli, T., and Zetzsche, C. 1993. Image encoding, labeling and reconstruction from differential geometry. CVGIP: Graphical Models and Image Processing, 55(6):428–446.Google Scholar
  7. Belhumeur, P. and Kriegman, D. 1996. What is the set of images of an object under all possible lighting conditions? In Proc. IEEE Conf. Computer Vision Pattern Recognition, San Francisco, IEEE Computer Society, IEEE Computer Society Press, pp. 270–277.Google Scholar
  8. Canny, J. 1983. Finding edges and lines in images. Master's thesis, MIT Artificial Intelligence Laboratory.Google Scholar
  9. Carlsson, S. 1984. Sketch based image coding. In Proc. of Premier Colloque Image, Biarritz, France, pp. 71–77.Google Scholar
  10. Carlsson, S. 1988. Sketch based coding of grey level images. Signal Processing, 15:57–83.Google Scholar
  11. Cox, I., Boie, R., and Wallach, D. 1990. Line recognition. In Proc. Int. Conf. on Pattern Recognition, Atlantic City, NJ, pp. 639–645.Google Scholar
  12. Cox, I., Rehg, J., and Hingorani, S. 1993. A Bayesian multiple-hypothesis approach to edge grouping and contour segmentation. Int. J. Comp. Vision, 11(1):5–24.Google Scholar
  13. Cumani, A., Grattoni, P., and Guiducci, A. 1991. An Edge-Based Description of Color Images. CVGIP: Graphical Models and Image Processing, 53(4):313–323.Google Scholar
  14. Curtis, S., Shitz, S., and Oppenheim, A. 1987. Reconstruction of nonperiodic two-dimensional signals from zero crossings. IEEE Trans. Acoust., Speech, Signal Processing, 35:890–893.Google Scholar
  15. Daubechies, I. 1991. Ten lectures on wavelets. In CBMS-NSF Series Appl. Math., SIAM.Google Scholar
  16. David, C. and Zucker, S. 1990. Potentials, valleys and dynamic global coverings. Int. J. Computer Vision, 5:219–238.Google Scholar
  17. Deriche, R. 1987. Using Canny's criteria to derive a recursively implemented optimal edge detector. Int. J. Computer Vision, 1(2):167–187.Google Scholar
  18. Dron, L. 1977. The multiscale veto model: a two-stage analog network for edge detection and image reconstruction. Int. J. Comp. Vision, 56:487–510.Google Scholar
  19. Elder, J. 1997. Brightness filling-in of natural images. European Conf. on Visual Perception, Helsinki, Finland, In Perception, 26(Suppl.) 57.Google Scholar
  20. Elder, J., Beniaminov, D., and Pintilie, G. 1999. Edge classification in natural images. J. Invest. Opthalm. Visual Sci. 40(4):1897.Google Scholar
  21. Elder, J. and Goldberg, R. 1998. Interactive contour editing. In Proc. IEEE Conf. Computer Vision Pattern Recognition, Santa Barbara, CA, IEEE Computer Society, IEEE Computer Society Press, pp. 374–381.Google Scholar
  22. Elder, J., Trithart, S., Pintilie, G., and MacLean, D. 1998. Rapid processing of cast and attached shadows. J. Invest. Opthalm. Visual Sci., 39(4):S853.Google Scholar
  23. Elder, J. and Zucker, S. 1995 The local character of generalized luminance transitions. J. Invest. Opthalm. Visual Sci., 36(4):S836.Google Scholar
  24. Elder, J. and Zucker, S. 1998. Local scale control for edge detection and blur estimation. IEEE Pattern Anal. Machine Intell., 20(7):699–716.Google Scholar
  25. Elder, J. and Zucker, S. 1996a. Computing contour closure. In Proc 4th European Conf. on Computer Vision, Lecture Notes in Computer Science, Springer Verlag, New York, pp. 399–412.Google Scholar
  26. Elder, J. and Zucker, S. 1996b. Local scale control for edge detection and blur estimation. In Proc. 4th European Conf. on Computer Vision, Lecture Notes in Computer Science, Springer Verlag, New York, pp. 57–69.Google Scholar
  27. Elder, J. and Zucker, S. 1996c. Scale space localization, blur and contour-based image coding. In Proc. IEEE Conf. Computer Vision Pattern Recognition, San Francisco. IEEE Computer Society, IEEE Computer Society Press, pp. 27–34.Google Scholar
  28. Ens, J. and Lawrence, P. 1993. Investigation of methods for determining depth from focus. IEEE Trans. Pattern Anal. Machine Intell., 15(2):97–108.Google Scholar
  29. Fischler, M., Tenenbaum, J., and Wolf, H. 1981. Detection of roads and linear structures in low-resolution aerial imagery using a multi-source knowledge integration technique. Computer Graphics and Image Processing, 18(4):201–223.Google Scholar
  30. Freeman, W. and Adelson, E. 1991. The design and use of steerable filters. IEEE Trans. Pattern Anal. Machine Intell., 13(9):891–906.Google Scholar
  31. Grattoni, P. and Guiducci, A. 1990. Contour Coding for Image Description. Pattern Recognition Letters, 11:95–105.Google Scholar
  32. Hummel, R. and Moniot, R. 1989. Reconstructions from zero crossings in scale space. IEEE Trans. on Acoustics, Speech, and Signal Processing, 37(12):2111–2130.Google Scholar
  33. Iverson, L. and Zucker, S. 1995. Logical/linear operators for image curves. IEEE Trans. Pattern Anal. Machine Intell., 17(10):982–996.Google Scholar
  34. Kersten, D., Mamassian, P., and Knill, D. 1997. Moving cast shadows induce apparent motion in depth. Perception, 26(2):171–192.Google Scholar
  35. Koenderink, J. 1984. The structure of images. Biol. Cybern., 50:363–370.Google Scholar
  36. Leclerc, Y. and Zucker, S. 1987. The local structure of image discontinuities in one dimension. IEEE Trans. Pattern Anal. Machine Intell., 9(3):341–355.Google Scholar
  37. Lindeberg, T. 1990. Scale-space for discrete signals. IEEE Trans. Pattern Anal. Machine Intell., 12(3):234–254.Google Scholar
  38. Lindeberg, T. 1996. Edge detection and ridge detection with automatic scale selection. In IEEE Conf. Computer Vision Pattern Recognition, San Francisco, IEEE Computer Society, IEEE Computer Society Press, pp. 465–470.Google Scholar
  39. Logan, B.F. 1977. Information in the zero-crossings of bandpass signals. Bell Syst. Tech. J., 56:487–510.Google Scholar
  40. Mallat, S. 1989. Multifrequency channel decompositions of images and wavelet models. IEEE Trans. Inform. Theory, 37(12):2091–2110.Google Scholar
  41. Mallat, S. and Zhong, S. 1992. Characterization of signals from multiscale edges. IEEE Trans. Pattern Anal. Machine Intell., 14:710–732.Google Scholar
  42. Marr, D. 1982. Vision. W.H. Freeman: New York.Google Scholar
  43. Marr, D. and Hildreth, E. 1980. Theory of edge detection. Proc. R. Soc. Lond. B, 207:187–217.Google Scholar
  44. Nayar, S. and Yasuo, N. 1994. Shape from focus. IEEE Trans. Pattern Anal. Machine Intell., 16(8):824–831.Google Scholar
  45. Paradiso, M. and Nakayama, K. 1991. Brightness perception and filling-in. Vision Res., 7/8:1221–1236.Google Scholar
  46. Pentland, A. 1987. A new sense for depth of field. IEEE Trans. Pattern Anal. Machine Intell., 9(4):523–531.Google Scholar
  47. Perona, P. 1995. Deformable kernels for early vision. IEEE Trans. Pattern Anal. Machine Intell., 17(5):488–499.Google Scholar
  48. Press, W., Teukolsky, S., Vetterling, W., and Flannery, B. 1992. Numerical Recipes in C (2 edition)., Cambridge University Press, chap 19, pp. 871–882.Google Scholar
  49. Roberts, L. 1965. Machine perception of 3-dimensional solids. In Optical and Electro-Optical Information Processing, J. Tippett (Ed.), MIT Press: Cambridge, MA.Google Scholar
  50. Sha'ashua, A. and Ullman, S. 1988. Structural saliency: the detection of globally salient structures using a locally connected network. In Proc. 2nd Int. Conf. on Computer Vision, Tampa, Florida. IEEE Computer Soc. Press, pp. 321–327.Google Scholar
  51. Simoncelli, E., Freeman, W., Adelson, E., and Heeger, D. 1992. Shiftable multiscale transforms. IEEE Trans. on Inf. Theory, 38(2):587–607.Google Scholar
  52. Vetterli, M. 1984. Multidimensional subband coding: some theory and algorithms. Signal Processing, 6(2):97–112.Google Scholar
  53. Werner, H. 1935. Studies on contour: I. Qualitative analyses. Amer. J. Psychol., 47:40–64.Google Scholar
  54. Witkin, A. 1983. Scale space filtering. In Proc. Int. Joint Conf. on Artif. Intell., Karlsruhe, pp. 1019–1021.Google Scholar
  55. Yuille, A. and Poggio, T. 1985. Fingerprints theorems for zero crossings. J. Opt. Soc. Am. A, 2(5):683–692.Google Scholar
  56. Zeevi, Y. and Rotem, D. 1986. Image reconstruction from zero crossings. IEEE Trans. Acoust., Speech, Signal Processing, 34:1269–1277.Google Scholar
  57. Zucker, S. 1986. Early vision. In The Encyclopedia of Artificial Intelligence, S. Shapiro (Ed.), John Wiley.Google Scholar
  58. Zucker, S., Hummel, R., and Rosenfeld, A. 1977. An application of relaxation labeling to line and curve enhancement. IEEE Trans. Comput., 26:394–403.Google Scholar

Copyright information

© Kluwer Academic Publishers 1999

Authors and Affiliations

  • James H. Elder

There are no affiliations available

Personalised recommendations