4.15 – Shape Smoothing and PDEs

One of the aims of Computer Vision in the past thirty years has been to recognize shapes with numerical algorithms. In this chapter, we describe five curve smoothing algorithms, of growing sophistication and invariance. We give a detailed implementation and link these algorithms to the curve evolution PDE's they implement. We let the five algorithms undergo a practical invariance testing. The tested invariance requirements face no less than five classes of perturbations, namely noise, geometric or affine distortions, contrast changes, occlusion and figure/background reversal. Quite contrary to the main stream idea that curve evolution schemes should be implemented by level set methods, we describe very fast and accurate direct curve evolution implementations. We give precise bibliographical links to the mathematical and image analysis literature justifying these curve evolution algorithms and their relationship to mathematical morphology, scale space theory, classical filtering. We finally give some hints on the role of shape smoothing in actual shape recognition systems.

[1]  Haim J. Wolfson,et al.  Model-Based Object Recognition by Geometric Hashing , 1990, ECCV.

[2]  King-Sun Fu,et al.  Shape Discrimination Using Fourier Descriptors , 1977, IEEE Trans. Syst. Man Cybern..

[3]  Yun-Gang Chen,et al.  Uniqueness and existence of viscosity solutions of generalized mean curvature flow equations , 1989 .

[4]  Farzin Mokhtarian,et al.  A Theory of Multiscale, Curvature-Based Shape Representation for Planar Curves , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Rachid Deriche,et al.  On corner and vertex detection , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  J. J. Koenderink,et al.  Dynamic shape , 1986, Biological Cybernetics.

[7]  P. Lions,et al.  Axioms and fundamental equations of image processing , 1993 .

[8]  Lionel Moisan,et al.  Edge Detection by Helmholtz Principle , 2001, Journal of Mathematical Imaging and Vision.

[9]  Farzin Mokhtarian,et al.  Scale-Based Description and Recognition of Planar Curves and Two-Dimensional Shapes , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  R. Bajcsy,et al.  A computerized system for the elastic matching of deformed radiographic images to idealized atlas images. , 1983, Journal of computer assisted tomography.

[11]  Roland T. Chin,et al.  On Image Analysis by the Methods of Moments , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  J. Sethian,et al.  Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations , 1988 .

[13]  David G. Lowe,et al.  Three-Dimensional Object Recognition from Single Two-Dimensional Images , 1987, Artif. Intell..

[14]  L. Evans.  Convergence of an algorithm for mean curvature motion , 1993 .

[15]  Sadegh Abbasi,et al.  Retrieval of Similar Shapes under Affine Transform , 1999, VISUAL.

[16]  Rama Chellappa,et al.  Classification of Partial 2-D Shapes Using Fourier Descriptors , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  L. Evans,et al.  Motion of level sets by mean curvature IV , 1995 .

[18]  Lionel Moisan.  Traitement numérique d'images et de films : équations aux dérivées partielles préservant forme et relief , 1997 .

[19]  F. Attneave.  Some informational aspects of visual perception. , 1954, Psychological review.

[20]  Jean-Michel Morel,et al.  Topographic Maps and Local Contrast Changes in Natural Images , 1999, International Journal of Computer Vision.

[21]  G. Sapiro,et al.  On the affine heat equation for non-convex curves , 1998 .

[22]  Paul Dupuis,et al.  Variational problems on ows of di eomorphisms for image matching , 1998 .

[23]  J. Morel,et al.  Connected components of sets of finite perimeter and applications to image processing , 2001 .

[24]  John K. Tsotsos,et al.  Shape Representation and Recognition from Multiscale Curvature , 1997, Comput. Vis. Image Underst..

[25]  Lionel Moisan,et al.  Affine plane curve evolution: a fully consistent scheme , 1998, IEEE Trans. Image Process..

[26]  M. Grayson.  The heat equation shrinks embedded plane curves to round points , 1987 .

[27]  Guillermo Sapiro,et al.  Affine invariant scale-space , 1993, International Journal of Computer Vision.

[28]  G. Kanizsa,et al.  Organization in Vision: Essays on Gestalt Perception , 1979 .

[29]  Adam Krzyzak,et al.  Reconstruction of two-dimensional patterns from Fourier descriptors , 2005, Machine Vision and Applications.

[30]  Max Wertheimer,et al.  Untersuchungen zur Lehre von der Gestalt , .

[31]  William J. Firey,et al.  Shapes of worn stones , 1974 .

[32]  Yehezkel Lamdan,et al.  Object recognition by affine invariant matching , 2011, Proceedings CVPR '88: The Computer Society Conference on Computer Vision and Pattern Recognition.

[33]  Shimon Ullman,et al.  Recognizing solid objects by alignment with an image , 1990, International Journal of Computer Vision.

[34]  Michael Brady,et al.  The Curvature Primal Sketch , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[35]  Lionel Moisan,et al.  Geometric Multiscale Representation of Numerical Images , 1999, Scale-Space.

[36]  G. Barles,et al.  A Simple Proof of Convergence for an Approximation Scheme for Computing Motions by Mean Curvature , 1995 .

[37]  Jitendra Malik,et al.  Matching Shapes , 2001, ICCV.

[38]  Jean-Michel Morel,et al.  Integral and local affine invariant parameter and application to shape recognition , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[39]  Jean-Michel Morel,et al.  Geometry and Color in Natural Images , 2002, Journal of Mathematical Imaging and Vision.

[40]  Lionel Moisan,et al.  On the Theory of Planar Shape , 2003, Multiscale Model. Simul..

[41]  Laurent Younes,et al.  Computable Elastic Distances Between Shapes , 1998, SIAM J. Appl. Math..

[42]  Hiroshi Murase,et al.  Learning and recognition of 3D objects from appearance , 1993, [1993] Proceedings IEEE Workshop on Qualitative Vision.