Quadratic Gabor filters for object detection

We present a new class of quadratic filters that are capable of creating spherical, elliptical, hyperbolic and linear decision surfaces which result in better detection and classification capabilities than the linear decision surfaces obtained from correlation filters. Each filter comprises of a number of separately designed linear basis filters. These filters are non-linearly combined using an algorithm called the extended piecewise quadratic neural network (E-PQNN). We consider the use of Gabor basis filters; the Gabor filter parameters are separately optimized; the fusion parameters used to combine the Gabor filter outputs are optimized using the conjugate gradient method; they and the non-linear combination of filter outputs are included in our E-PQNN algorithm. We present results obtained for an infra-red (IR) vehicle detection problem.

[1]  M. J. D. Powell,et al.  Restart procedures for the conjugate gradient method , 1977, Math. Program..

[2]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[3]  Etienne Barnard,et al.  A comparison between criterion functions for linear classifiers, with an application to neural nets , 1989, IEEE Trans. Syst. Man Cybern..

[4]  Keinosuke Fukunaga,et al.  Introduction to statistical pattern recognition (2nd ed.) , 1990 .

[5]  B. MacLennan.  Gabor Representations of Spatiotemporal Visual Images , 1991 .

[6]  D. Casasent,et al.  Gaussian-minimum average correlation energy filters. , 1991, Applied optics.

[7]  N. Ranganathan,et al.  Gabor filter-based edge detection , 1992, Pattern Recognit..

[8]  B V Kumar,et al.  Tutorial survey of composite filter designs for optical correlators. , 1992, Applied optics.

[9]  David Casasent,et al.  New basis function distortion invariant detection filters , 1994, Optics & Photonics.

[10]  David Casasent,et al.  Real, imaginary, and clutter Gabor filter fusion for detection with reduced false alarms , 1994 .

[11]  David Casasent,et al.  Neural net design of macro Gabor wavelet filters for distortion-invariant object detection in clutter , 1994 .

[12]  J. Shewchuk.  An Introduction to the Conjugate Gradient Method Without the Agonizing Pain , 1994 .

[13]  Darren Redfern,et al.  The Maple handbook , 1994 .

[14]  Gregory O. Gheen.  General class of invariant quadratic filters for optical pattern recognition , 1994, Defense, Security, and Sensing.

[15]  David Casasent,et al.  Fusion and optimized Gabor filter design for object detection , 1995, Other Conferences.

[16]  David Casasent,et al.  A classifier neural net with complex-valued weights and square-law nonlinearities , 1995, Neural Networks.

[17]  David A. Redfern,et al.  The Maple Handbook: Maple V Release 3 , 1995 .

[18]  L. Råde,et al.  Mathematics handbook for science and engineering , 1995 .

[19]  David Casasent.  Gabor wavelet filters and fusion for distortion-invariant multiclass object detection , 1995, Defense, Security, and Sensing.

[20]  David Casasent,et al.  Detection filters and algorithm fusion for ATR , 1997, IEEE Trans. Image Process..

[21]  J. Breitenbach,et al.  Neural network objective functions for detection problems , 1997, Proceedings of the 1997 South African Symposium on Communications and Signal Processing. COMSIG '97.

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

[23]  Rajesh Shenoy,et al.  Synthetic aperture radar detection and clutter rejection minace filters , 1997, Pattern Recognit..

[24]  David M. Weber,et al.  The extended piecewise quadratic neural network , 1998, Neural Networks.