Higher-degree implicit polynomials and moments provide useful global descriptors for complex curves and surfaces. In this paper, a tensor-based approach for obtaining Euclidean and affine invariants from coefficients of higher-degree implicit polynomials has been pro posed. The algorithm is an extension of the matrix-based approach by Taubin, but unlike the previous works, it is not based on partial derivative forms or symbolic computation. Owing to the close re lationship between algebraic invariants and moment invariants, the approach is equally applicable to moment invariants.
Get full access to this article
View all access options for this article.
References
1.
Abhyankar, S.S.1992. Invariant theory and enumerative combinatorics of young tableaux. Geometric Invariance in Computer Vision, eds. J. L. Mundy and A. Zisserman. Cambridge, MA: MIT Press, pp. 45-76.
2.
Bookstein, F.L.1979. Fitting conic sections to scattered data. CompGraphics and Image Proc. 9:56-71.
3.
Brown, W.C.1991. Matrices and Vector SpacesNew York : Marcel Dekker.
4.
Cyganski, D., and Orr, J.A.1985. Applications of tensor theory to object recognition and orientation determinationIEEE Trans. Pattern Analysis Machine Intell.7(6):662-673.
5.
Gool, L.V., Moons, T., Pauwels, E., and Oosterlinck, A.1995. Vision and Lie's approach to invarianceImage and Vision Computing13(4):259-277.
6.
Grimson, W.E.L., and Huttenlocher, D.P.1990 (Osaka). On the sensitivity of geometric hashing. Proc. of the Third Int. Conf. on Comp. VisionLos Alamitos, CA: IEEE, pp. 334-338.
7.
Hu, M.K.1962. Visual pattern recognition by moment invariantsIRE Trans. Information Theory8(2):179-187.
8.
Kaveti, S., Teoh, E.K., and Wang, H.1997. Efficient algo rithms for obtaining algebraic invariants from higher degree implicit polynomials for recognition of curved objectsPattern Recognition J.31(3):301-313.
9.
Keren, D.1994. Using symbolic computation to find algebraic invariantsIEEE Trans. Pattern Analysis Machine Intell.16(11):1143-1149.
10.
Keren, D., Cooper, D., and Subramonia, J.1994. Describing complicated objects by implicit polynomialsIEEE Trans. Pattern Analysis Machine Intell.16(1):38-53.
11.
Knuth, D.E.1968. Fundamental Algorithms, vol. 1 of The Art of Computer ProgrammingReading, MA: Addison-Wesley .
12.
Lamdan, Y., and Wolfson, J.J.1988 (Tarpon Springs, FL). Geometric hashing: A general and efficient model-based recognition scheme. Proc. of the Second Int. Conf on Comp. VisionLos Alamitos, CA: IEEE, pp. 238-249.
13.
Lamdan, Y., and Wolfson, H.J.1991 (Maui, HI). On the error analysis of geometric hashing. Proc. of the CVPRLos Alamitos, CA: IEEE, pp. 22-27.
14.
Lawson, C.L., and Hanson, R.J.1974. Solving Least Squares ProblemsEnglewood Cliffs, NJ: Prentice-Hall.
15.
Lovelock, D., and Rund, H.1975. Tensors, Differential Forms, and Variational PrinciplesNew York: Wiley Interscience.
16.
Moons, T., Pauwels, E.J., Gool, L. J. V., and Oosterlinck, A.1995. Foundations of semi-differential invariantsInt. J. Comp. Vision14:25-47.
17.
Parshall, K.H.1989. Toward a history of nineteenth-century invariant theory. The History of Modern MathematicsNew York: Academic.
18.
Pei, S.-C., and Lin, C.-N.1995. Image normalization for pattern recognitionImage and Vision Computing13(10):711-723.
19.
Pratt, V.1987. Direct least squares fitting of algebraic surfacesComp. Graphics21(4):145-152.
20.
Sampson, P.D.1982. Fitting conic sections to very scattered data: An iterative refinement of the Bookstein algorithmComp. Graphics Image Proc.18:97-108.
21.
Spiegal, M.R.1974. Schaum's Outline of Theory and Problems of Vector Analysis and an Introduction to Tensor AnalysisNew York : McGraw-Hill.
22.
Strumfels, B.1993. Algorithms in Invariant TheoryNew York: Springer-Verlag.
23.
Subramonia, J., Keren, D., and Cooper, D.B.1993 (Berlin). Recognizing mice, vegetables, and hand-printed characters based on implicit polynomials, invariants, and Bayesian methods. Proc. Fourth Int. Conf. on Comp. VisionLos Alamitos, CA : IEEE, pp. 320-324.
24.
Sullivan, S., Sandford, L., and Ponce, J.1994. Using geometric distance fits for 3-D object modeling and recognitonIEEE Trans. Pattern Analysis Machine Intell.16(12):1183-1196.
25.
Synge, J.L., and Shild, A.1978. Tensor Calculus. New York: Dover.
26.
Taubin, G.1990. Recognition and positioning of rigid objects using algebraic and moment invariants Technical Report LEMS-80, Brown University, Providence, RI.
27.
Taubin, G.1991. Estimation of planar curves, surfaces, and nonplanar space curves defined by implicit equations with applications to edge and range image segmentationIEEE Trans. Pattern Analysis Machine Intell.13(11):1115-1138.
28.
Taubin, G., Cukierman, F., Sullivan, S., Ponce, J., and Kriegman, D.J.1994. Parameterized families of polynomials for bounded algebraic curve and surface fittingIEEE Trans. Pattern Analysis Machine Intell. 16(3):287-303.
