Decision regions of Fourier-plane nonlinear filtering for image recognition

In image recognition applications, complex decision regions in the image space are needed. Linear filtering forms the decision regions by hyperplanes in the image space. We determine the decision region formed by Fourier-plane nonlinear filtering. In the case in which power law nonlinearity is applied in the Fourier plane, the decision region turns out to be approximately an n-dimensional parabola that opens toward the direction of the reference vector. That is, the intersection of the decision region with any plane (two-dimensional vector space) not containing any vector parallel to the reference vector is a bounded convex region enclosed by a closed curve. The size of the convex region depends on the filter nonlinearity, which determines the distortion robustness and discrimination capability of the filter. It can be adjusted by choosing different Fourier-plane nonlinearities and/or different threshold values at the output plane. These types of regions are desirable and well suited in image recognition. Analytical and numerical solutions are provided.

[1]  Bahrain Javidi,et al.  Performance of the nonlinear joint transform correlator for images with low-pass characteristics. , 1994, Applied optics.

[2]  David L. Flannery,et al.  Design elements of binary joint transform correlation and selected optimization techniques , 1992 .

[3]  Philippe Réfrégier,et al.  DECISION THEORY APPROACH TO NONLINEAR JOINT-TRANSFORM CORRELATION , 1998 .

[4]  Demetri Psaltis,et al.  Optical-Image Correlation With A Binary Spatial Light-Modulator , 1984 .

[5]  D Psaltis,et al.  Optical network for real-time face recognition. , 1993, Applied optics.

[6]  Bahram Javidi,et al.  Binary nonlinear joint transform correlator performance with different thresholding methods under unknown illumination conditions. , 1995, Applied optics.

[7]  B Javidi,et al.  Optical implementation of neural networks for face recognition by the use of nonlinear joint transform correlators. , 1995, Applied optics.

[8]  David Casasent,et al.  Quadratic filters for object classification and detection , 1997, Defense, Security, and Sensing.

[9]  B. Javidi.  Nonlinear joint power spectrum based optical correlation. , 1989, Applied optics.

[10]  B Javidi,et al.  Nonlinear joint-transform correlation: an optimal solution for adaptive image discrimination and input noise robustness. , 1994, Optics letters.

[11]  J. Horner,et al.  Phase-only matched filtering. , 1984, Applied optics.

[12]  H J Caulfield,et al.  Improved discrimination in optical character recognition. , 1969, Applied optics.

[13]  P. Réfrégier.  Filter design for optical pattern recognition: multicriteria optimization approach. , 1990, Optics letters.

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

[15]  Bahram Javidi,et al.  Distortion-invariant pattern recognition with Fourier-plane nonlinear filters. , 1996, Applied optics.

[16]  Joseph L. Horner,et al.  1-F Binary Joint Transform Correlator. , 1990 .