On the minimum number of templates required for shift, rotation and size invariant pattern recognition

Abstract Human observers are generally capable of recognizing patterns invariant to their orientation, position and size within an image. Though techniques are available for similar performance in computer visual systems, most suffer from lack of uniqueness or computational complexity. In this paper we introduce a new adaptive approach to invariant pattern recognition which overcomes both these problems. This technique is based upon the intrinsic invariance properties of the pattern and the recognition criterion. Our simulations demonstrate that the number of templates required to gain efficient pattern recognition is considerably lower than previously thought.

[1]  D. Casasent,et al.  Position, rotation, and scale invariant optical correlation. , 1976, Applied optics.

[2]  A. Cormack.  Representation of a Function by Its Line Integrals, with Some Radiological Applications , 1963 .

[3]  H Ghandeharian,et al.  Visual signal detection. I. Ability to use phase information. , 1984, Journal of the Optical Society of America. A, Optics and image science.

[4]  H Stark,et al.  Rotation-invariant pattern recognition using a vector reference , 1984 .

[5]  Donald W. Sweeney,et al.  Iterative technique for the synthesis of optical-correlation filters , 1986 .

[6]  D. Casasent,et al.  New optical transforms for pattern recognition , 1977, Proceedings of the IEEE.

[7]  H H Arsenault,et al.  Optical pattern recognition using circular harmonic expansion. , 1982, Applied optics.

[8]  B. V. K. Vijaya Kumar,et al.  Loss of optimality in cross correlators , 1984 .

[9]  Henri H. Arsenault,et al.  Optimum Circular Symmetrical Filters and Their Uses in Optical Pattern Recognition , 1982 .

[10]  Azriel Rosenfeld,et al.  Digital Picture Processing , 1976 .

[11]  H. Caulfield,et al.  Computer recognition of 2-D patterns using generalized matched filters. , 1982, Applied optics.

[12]  Terry Caelli,et al.  Fast edge-only matching techniques for robot pattern recognition , 1987 .

[13]  H H Arsenault,et al.  Rotation-invariant digital pattern recognition using circular harmonic expansion. , 1982, Applied optics.

[14]  B. V. Vijaya Kumar,et al.  Minimum-variance synthetic discriminant functions , 1986 .

[15]  D Casasent,et al.  Unified synthetic discriminant function computational formulation. , 1984, Applied optics.

[16]  D Casasent,et al.  Polar camera for space-variant pattern recognition. , 1978, Applied optics.