This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Deformable Kernels for Early Vision
May 1995 (vol. 17 no. 5)
pp. 488-499

Abstract—Early vision algorithms often have a first stage of linear-filtering that ’extracts’ from the image information at multiple scales of resolution and multiple orientations. A common difficulty in the design and implementation of such schemes is that one feels compelled to discretize coarsely the space of scales and orientations in order to reduce computation and storage costs. This discretization produces anisotropies due to a loss of translation-, rotation-, and scaling-invariance that makes early vision algorithms less precise and more difficult to design. This need not be so: one can compute and store efficiently the response of families of linear filters defined on a continuum of orientations and scales. A technique is presented that allows 1) computing the best approximation of a given family using linear combinations of a small number of ’basis’ functions; 2) describing all finite-dimensional families, i.e., the families of filters for which a finite dimensional representation is possible with no error. The technique is based on singular value decomposition and may be applied to generating filters in arbitrary dimensions and subject to arbitrary deformations; the relevant functional analysis results are reviewed and precise conditions for the decomposition to be feasible are stated. Experimental results are presented that demonstrate the applicability of the technique to generating multi-orientation multi-scale 2D edge-detection kernels. The implementation issues are also discussed.

[1] D. Hubel and T. Wiesel,“Receptive fields, binocular interaction and functional architecture inthe cat’s visual cortex.,” J. Physiol. (Lond.), vol. 160, pp. 106-154, 1962.
[2] D. Hubel and T. Wiesel,“Receptive fields of single neurones in the cat’s striate cortex.,” J. Physiol. (Lond.), vol. 148, pp. 574-591, 1959.
[3] B. Horn,“The binford-horn linefinder,” Technical report, MIT AI Lab. Memo 285, 1971.
[4] T. Binford,“Inferring surfaces from images,” Artificial Intelligence, vol. 17, pp. 205-244, 1981.
[5] G.H. Granlund,“In search of a general picture processing operator,” Computer Graphics and Image Processing, vol. 8, pp. 155-173, 1978.
[6] H. Knuttson and G.H. Granlund,“Texture analysis using two-dimensional quadrature filters,” Workshop on Computer Architecture for Pattern Analysis and Image Database Management, pp. 206-213, IEEE Computer Society, 1983.
[7] M. Kass,“Computing visual correspondence,” Proc. Image Understanding Workshop,(McLean, Va.), pp. 54-60, Science Applications, Inc., June 1983.
[8] P.J. Burt and E.H. Adelson, “The Laplacian Pyramid as a Compact Image Code,” IEEE Trans. Comm., vol. 31, no. 4, pp. 532-540, 1983.
[9] E. Adelson and J. Bergen,“Spatiotemporal energy models for the perception of motion,” J. Opt. Soc. Am., vol. 2, no. 2, pp. 284-299, 1985.
[10] J. Canny, “A Computational Approach to Edge Detection,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 8, no. 6, pp. 679-698, June 1986.
[11] D. Heeger,“Optical flow from spatiotemporal filters,” Proc. First Int’l Conf. Computer Vision, pp. 181-190, 1987.
[12] M. Turner,“Texture discrimination by gabor functions,” Biol. Cybern., vol. 55, pp. 71-82, 1986.
[13] M.C. Morrone and R.A. Owens, "Feature Detection From Local Energy," Pattern Recognition Letters, vol. 6, pp. 303-313, 1987.
[14] M. Morrone and D. Burr,“A model of human feature detection based on matched filters,” Robots and Biological Systems, P. Dario and G. Sandini, ed., Academic Press, 1990.
[15] P. Parent and S.W. Zucker, “Trace Inference, Curvature Consistency and Curve Detection,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, no. 8, pp. 823-839, 1989.
[16] I. Fogel and D. Sagi,“Gabor filters as texture discriminators,” Biol. Cybern., vol. 61, pp. 103-113, 1989.
[17] J. Malik and P. Perona,“Preattentive texture discrimination with early vision mechanisms,” J. Opt. Soc. Am. A, vol. 7, no. 5, pp. 923-932, 1990. Also appeared as a chapter in Computer vision: advances and applications, R. Kasturi and R. Jain, ed., IEEE Computer Society Press, 1991.
[18] A.C. Bovik,M. Clark,, and W.S. Geisler,“Multichannel texture analysis using localized spatial filters,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 12, pp. 55-73, 1990.
[19] S. Zhong and S. Mallat,“Compact image representation from multiscale edges,” Proc. Third Int’l Conf. Computer Vision, IEEE Computer Society, Osaka, 1990.
[20] D. Jones and J. Malik,“Computational stereopsis-Beyond zero-crossings,” Invest. Ophtalmol. Vis. Sci. (Supplement), vol. 31, no. 4, p. 529, 1990.
[21] D. Jones and J. Malik,“Using orientation and spatial frequency disparities to recover 3d surface shape—A computational model,” Invest. Ophtalmol. Vis. Sci. (Supplement), vol. 32, no. 4, p. 710, 1991.
[22] W. T. Freeman and E. H. Adelson,“Steerable filters for early vision, image analysis, and wavelet decomposition,” Proc. of ICCV, pp. 406-415,Osaka, Japan, 1990.
[23] P. Perona and J. Malik, "Detecting and Localizing Edges Composed of Steps, Peaks and Roofs," Proc. Third Int'l Conf. Computer Vision,Osaka, Japan, pp. 52-57, 1990.
[24] E. Adelson and J. Bergen,“The plenoptic function and the elements of early vision,” Computational Models of Visual Processing, M. Landy and J. Movshon, ed., MIT Press, 1991. Also appeared as MIT-MediaLab-TR148. Sept. 1990.
[25] J. Koenderink and A. van Doorn,“Representation of local geometry in the visual system,” Biol. Cybern., vol. 55, pp. 367-375, 1987.
[26] J. Koenderink and A. van Doorn,“Receptive field families,” Biol. Cybern., vol. 63, pp. 291-298, 1990.
[27] R. Lenz, Group Theoretical Methods in Image Processing. New York: Springer-Verlag, 1990.
[28] B.M. ter Haar Romeny,L.M.J. Florack,J.J. Koenderink,, and M.A. Viergever,“Scale-space: its natural operators and differential invariants,” Int’l Conf. Information Processing in Medical Imaging, vol. 511, Lecture Notes in Computer Sciences, pp. 239-255,Springer-Verlag, 1992.
[29] J. Daugman,“Uncertainty relation for resolution in space, spatial frequency, andorientation optimized by two-dimensional visual cortical filters,” J. Opt. Soc. Am. A, vol. 2, no. 7, pp. 1160-1169, 1985.
[30] J. Daugman,“Two-dimensional spectral analysis of cortical receptive field profiles,” Vis. Research, vol. 20, pp. 847-856, 1980.
[31] R. Young,“The gaussian derivative theory of spatial vision: Analysis of corticalcell receptive field line-weighting profiles,” Technical Report GMR-4920, General Motors Research, 1985.
[32] M. Morrone and D. Burr,“Feature detection in human vision: a phase dependent energy model,” Proc. R. Soc. Lond. B, vol. 235, pp. 221-245, 1988.
[33] D. Jones and J. Malik, “A Computational Framework for Determining Stereo Correspondences from a Set of Linear Spatial Filters,” Proc. Second European Conf. Computer Vision, pp. 395-410, 1992.
[34] D. Jones and J. Malik,“Determining 3d shape from orientation and spatial frequncy disparitiesii—Using correspondence image patches,” UCB-CSD 91-657, U.C.Berkeley—CS Dept., Oct, 1991.
[35] J. Malik and Z. Gigus,“A model for curvilinear segregation,” Invest. Ophtalmol. Vis. Sci. (Supplement), vol. 32, no. 4, p. 715, 1991.
[36] L. Rosenthaler,F. Heitger,R. von der Heydt,, and O. Kübler,“Detecting general edges and keypoints,” Tech. Rep. 130, IKT/Image science lab ETH-Zuerich, 1991.
[37] F. Heitger,L. Rosenthaler,R. von der Heydt,E. Peterhans,, and O. Kübler,“Simulation of neural contour mechanismn: From single to end-stopped cell,” Vis. Research, vol. 32, no. 5, pp. 963-981, 1992.
[38] J.G. Daugman, “Complete Discrete 2D Gabor Transforms by Neural Networks for Image Analysis and Compression,” IEEE Trans. Acoustics, Speech, and Signal Processing, vol. 36, no. 7, 1988.
[39] M. Morrone,D. Burr,J. Ross,, and R. Owens,“Mach bands depend on spatial phase,” Nature, vol. 324, no. 6094, pp. 250-253, 1986.
[40] R. Hummel,“Feature detection using basis functions,” Comp. Vision, Graphics and Image Proc., vol. 9, pp. 40-55, 1979.
[41] H. Knuttson,Filtering and reconstruction in image processing. PhD thesis, Linköping Inst. of Tech nology, 1982.
[42] J. Bigün, G. Granlund, and J. Wiklund, “Multidimensional Orientation Estimation with Applications to Texture Analysis and Optical Flow,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, no. 8, pp. 775-790, Aug. 1991.
[43] W. Freeman and E. Adelson,“Steerable filters,” Topical Meeting on Image Understanding and Machine Vision, vol. 14, Optical Soc. America, technical digest series, Cape Cod, June 1989.
[44] W. Freeman and E. Adelson,“Steerable filters for image analysis,” Technical Report 126, MIT, Media Laboratory, 1990.
[45] W.T. Freeman and E.H. Adelson, "The Design and Use of Steerable Filters," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, pp. 891-906, 1991.
[46] J. Dworak et al., "Enhanced DO-RE-ME Based Defect Level Prediction Using Defect Site Aggregation—MPG-D," Proc. Int'l Test Conf., IEEE CS Press, Los Alamitos, Calif., pp. 930-939.
[47] P. Perona and J. Malik,“Boundary detection using quadratic filters: performance criteria andexperimental assessment.,” Proc. SPIE Conf. Applications of Artificial Intelligence, vol. 1708, Orlando, April 1992.
[48] E.P. Simoncelli, W.T. Freeman, E.H. Adelson, and D.J. Heeger, “Shiftable Multi-Scale Transforms,” IEEE Trans. Information Theory, vol. 38, no. 2, pp. 587-607, Mar. 1992.
[49] P. Perona,“Finite representation of deformable functions,” Tech. Rep. 90-034, Int’l Computer Science Inst., Berkeley, Calif., 1990.
[50] P. Perona,“Deformable kernels for early vision,” Proc. IEEE CS Conf. Computer Vision and Pattern Recognition, pp. 222-227,Maui, June 1991.
[51] P. Perona,“Deformable kernels for early vision,” Tech. Rep. 2039, LIDS-MIT, Oct. 1991.
[52] P. Perona,“Steerable-scalable kernels for edge detection and junctionnalysis,” Second European Conf. Computer Vision, vol. 588, Lecture Notes in Computer Sciences, Springer-Verlag, 1992.
[53] G. Choquet,Lectures on Analysis, vol. I. New York: W. A. Benjamin Inc., 1969.
[54] J. Dieudonne,Foundations of Modern Analysis.New York: Academic Press, 1969.
[55] A. Pinkus,n-Widths in Approximation Theory. Springer Verlag, 1985.
[56] S. Treitel and J. Shanks,“The design of multistage separable planar filters,” IEEE Trans. Geosci. Electron., vol. 9, no. 1, pp. 10-27, 1971.
[57] W.K. Pratt,“An intelligent image processing display terminal,” Proc. SPIE Tech. Symp., vol 27, San Diego, 1979.
[58] R. Deriche and J.-F. Abramatic.,“Design of 2d recursive filters using singular value decomposition techniques,” Proc. ICASSP, 1982.
[59] J. Dongarra,C. Moler,J. Bunch,, and G. Stuart,Linpack, User’s Guide. Soc. Industrial and Applied Mathematics, 1972.
[60] W. Press,B. Flannery,S. Teukolsky,, and W. Vetterling,Numerical Recipes in C. Cambridge University Press, 1988.
[61] J.-F. Abramatic and O. Faugeras,“Sequential convolution techniques for image filtering,” IEEE Trans. Acoustics, Speech, Signal Processing, vol. 30, no. 1, pp. 1-10, 1982.
[62] E.P. Simoncelli, W.T. Freeman, E.H. Adelson, and D.J. Heeger, “Shiftable Multi-Scale Transforms,” IEEE Trans. Information Theory, vol. 38, no. 2, pp. 587-607, Mar. 1992.
[63] M.H. Hueckel, An Operator which Locates Edges in Digitized Pictures J. ACM, vol. 18, no. 1, pp. 113-125, Jan. 1971.
[64] M.H. Hueckel, A Local Visual Operator which Recognizes Edges and Lines J. ACM, vol. 20, no. 4, pp. 634-647, 1973.
[65] R. Haralik,“Digital step edges from zero crossing of second directional derivatives,” IEEE Trans. Pattern Analysis and Machine Intelligence., vol. 6, no. 1, pp. 58-68, 1984.
[66] W. Rudin,Functional Analysis. McGraw-Hill, 1973.

Index Terms:
Steerable filters, wavelets, early vision, multiresolution image analysis, multirate filtering, deformable filters, scale-space
Citation:
Pietro Perona, "Deformable Kernels for Early Vision," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, no. 5, pp. 488-499, May 1995, doi:10.1109/34.391394
Usage of this product signifies your acceptance of the Terms of Use.