# A study of affine matching with bounded sensor error

## Abstract

Affine transformations of the plane have been used by modelbased recognition systems to approximate the effects of perspective projection. Because the underlying mathematics are based on *exact* data, in practice various heuristics are used to adapt the methods to real data where there is positional uncertainty. This paper provides a precise analysis of affine point matching under uncertainty. We obtain an expression for the range of affine-invariant values consistent with a given set of four points, where each data point lies in an ∃-disc. This range is shown to depend on the actual *x- y*-positions of the data points. Thus given uncertainty in the data, the representation is no longer invariant with respect to the Cartesian coordinate system. This is problematic for methods, such as geometric hashing, that depend on the invariant properties of the representation. We also analyze the effect that uncertainty has on the probability that recognition methods using affine transformations will find false positive matches. We find that such methods will produce false positives with even moderate levels of sensor error.

## Keywords

Object Recognition Image Point Recognition Method Model Point Affine Transformation## References

- 1.Ballard, D.H., 1981, “Generalizing the Hough Transform to Detect Arbitrary Patterns,”
*Pattern Recognition***13**(2): 111–122.CrossRefGoogle Scholar - 2.Basri, R. & S. Ullman, 1988, “The Alignment of Objects with Smooth Surfaces,”
*Second Int. Conf. Comp. Vision*, 482–488.Google Scholar - 3.Besl, P.J. & R.C. Jain, 1985, “Three-dimensional Object Recognition,”
*ACM Computing Surveys*,**17**(1):75–154.CrossRefGoogle Scholar - 4.Costa, M., R.M. Haralick & L.G. Shapiro, 1990, “Optimal Affine-Invariant Point Matching,”
*Proc. 6th Israel Conf. on AI*, pp. 35–61.Google Scholar - 5.Cyganski, D. & J.A. Orr, 1985, “Applications of Tensor Theory to Object Recognition and Orientation Determination”,
*IEEE Trans. PAMI*,**7**(6):662–673.Google Scholar - 6.Efimov, N.V., 1980,
*Higher Geometry*, translated by P.C. Sinha. Mir Publishers, Moscow.Google Scholar - 7.Ellis, R.E., 1989, “Uncertainty Estimates for Polyhedral Object Recognition,”
*IEEE Int. Conf. Rob. Aut.*, pp. 348–353.Google Scholar - 8.Forsythe, D., J.L. Mundy, A. Zisserman, C. Coelho, A. Heller, & C. Rothwell, 1991, “Invariant Descriptors for 3-D Object Recognition and Pose”,
*IEEE Trans. PAMI*,**13**(10):971–991.Google Scholar - 9.Grimson, W.E.L., 1990,
*Object Recognition by Computer: The role of geometric constraints*, MIT Press, Cambridge.Google Scholar - 10.Grimson, W.E.L., 1990, “The Combinatorics of Heuristic Search Termination for Object Recognition in Cluttered Environments,”
*First Europ. Conf. on Comp. Vis.*, pp. 552–556.Google Scholar - 11.Grimson, W.E.L. & D.P. Huttenlocher, 1990, “On the Sensitivity of the Hough Transform for Object Recognition,”
*IEEE Trans. PAMI***12**(3):255–274.Google Scholar - 12.Grimson, W.E.L. & D.P. Huttenlocher, 1991, “On the Verification of Hypothesized Matches in Model-Based Recognition”,
*IEEE Trans. PAMI***13**(12):1201–1213.Google Scholar - 13.Grimson, W.E.L. & D.P. Huttenlocher, 1990, “On the Sensitivity of Geometric Hashing”,
*Proc. Third Int. Conf. Comp. Vision*, pp. 334–338.Google Scholar - 14.Grimson, W.E.L., D.P. Huttenlocher, & D.W. Jacobs, 1991, “Affine Matching With Bounded Sensor Error: A Study of Geometric Hashing & Alignment,” MIT AI Lab Memo 1250.Google Scholar
- 15.Huttenlocher, D.P. and S. Ullman, 1987, “Object Recognition Using Alignment”,
*Proc. First Int. Conf. Comp. Vision*, pp. 102–111.Google Scholar - 16.Huttenlocher, D.P. & S. Ullman, 1990, “Recognizing Solid Objects by Alignment with an Image,”
*Inter. Journ. Comp. Vision***5**(2):195–212.Google Scholar - 17.Jacobs, D., 1991, “Optimal Matching of Planar Models in 3D Scenes,”
*IEEE Conf. Comp. Vis. and Patt. Recog.*pp. 269–274.Google Scholar - 18.Klein, F., 1939,
*Elementary Mathematics from an Advanced Standpoint: Geometry*, MacMillan, New York.Google Scholar - 19.Korn, G.A. & T.M. Korn, 1968,
*Mathematical Handbook for Scientists and Engineers*, McGraw-Hill, New York.Google Scholar - 20.Lamdan, Y., J.T. Schwartz & H.J. Wolfson, 1988, “Object Recognition by Affine Invariant Matching,”
*IEEE Conf. Comp. Vis. and Patt. Recog.*pp. 335–344.Google Scholar - 21.Lamdan, Y., J.T. Schwartz & H.J. Wolfson, 1990, “Affine Invariant Model-Based Object Recognition,”
*IEEE Trans. Rob. Aut.*, vol. 6, pp. 578–589.Google Scholar - 22.Lamdan, Y. & H.J. Wolfson, 1988, “Geometric Hashing: A General and Efficient Model-Based Recognition Scheme,”
*Second Int. Conf. Comp. Vis.*pp. 238–249.Google Scholar - 23.Lamdan, Y. & H.J. Wolfson, 1991, “On the Error Analysis of ‘Geometric Hashing',”
*IEEE Conf. Comp. Vis. and Patt. Recog.*pp. 22–27.Google Scholar - 24.Thompson, D. & J.L. Mundy, 1987, “Three-Dimensional Model Matching From an Unconstrained Viewpoint”,
*Proc. IEEE Conf. Rob. Aut.*pp. 280.Google Scholar - 25.Van Gool, L., P. Kempenaers & A. Oosterlinck, 1991, “Recognition and Semi-Differential Invariants,”
*IEEE Conf. Comp. Vis. and Patt. Recog.*pp. 454–460.Google Scholar - 26.Wayner, P.C., 1991, “Efficiently Using Invariant Theory for Model-based Matching,”
*IEEE Conf. Comp. Vis. and Patt. Recog.*pp. 473–478.Google Scholar - 27.Weiss, I., 1988, “Projective Invariants of Shape,”
*DARPA IU Workshop*pp. 1125–1134.Google Scholar - 28.Wolfson, H.J., 1990, “Model Based Object Recognition by Geometric Hashing,” First
*Europ. Conf. Comp. Vis.*pp. 526–536.Google Scholar