BIOLOGICALLY INSPIRED FILTERS UTILIZING SPECTRAL PROPERTIES OF TOEPLITZ-BLOCK-TOEPLITZ MATRICES
DOI:
https://doi.org/10.47839/ijc.14.4.820Keywords:
Lateral Inhibition Excitation, Recurrent Neural Networks, Toeplitz-Block-Toeplitz Matrices.Abstract
The construction of filters arising from linear neural networks with feed-backward excitatory-inhibitory connections is presented. Spatially invariant coupling between neurons and the distribution of neuron-receptor units in the form of a uniform square grid yield the TBT (Toeplitz-Block-Toeplitz) connection matrix. Utilizing the relationship between spectral properties of such matrices and their generating functions, the method for construction of recurrent linear networks is addressed. By appropriately bounding the generating function, the connection matrix eigenvalues are kept in the desired range allowing for large matrix inverse to be approximated by a convergent power series. Instead of matrix inversion, the single pass convolution with the filter obtained from the network connection weights is applied when solving the network. For the case of inter-neuron coupling in the form of a function that is expandable in a Fourier series in polar angle, the network response filter is shown to be steerable.References
H.K. Hartline, F. Ratliff, Inhibitory interaction of receptor units in the eye of Limulus, J. Gen. Physiol., (40) 3 (1957), pp. 357-376.
H.K. Hartline, F. Ratliff, Spatial summation of inhibitory influences in the eye of Limulus, and the mutual interaction of receptor units, J. Gen. Physiol., (41) 5 (1958), pp. 1049-1066.
B.S. Gutkin, C.E. Smith, Conditions for noise reduction and stable encoding of spatial structure by cortical neural networks, Biol. Cybern., (82) 6 (2000), pp. 469-475.
G.G. Furman, Comparison of models for subtractive and shunting lateral-inhibition in receptor-neuron fields, Biol. Cybern., (2) 6 (1965), pp. 257-274.
R.A. Messner, Smart Visual Sensors for Real-Time Image Processing and Pattern Recognition Based Upon Human Visual System Characteristics, PhD Dissertation, Clarkson University, 1984.
H.H. Szu, R.A. Messner, Adaptive invariant novelty filters, Proc. IEEE, (74) 3 (1986), pp. 518-519.
S.E. Palmer, Vision Science Photons to Phenomenology, A Bradford Book, The MIT Press, Cambridge, Massachusetts, London, England, 1999.
M.G. Luniewicz, R.A. Messner, Effects of lateral subtractive inhibition within the context of a polar-log spatial coordinate mapping, in: D.P. Casasent (Ed.), Proc. SPIE Intelligent Robots and Computer Vision VII, 1988, vol. 1002, pp. 58-65.
Y. Yu, T. Yamauchi, Y. Choe, Explaining low-level brightness-contrast illusions using disinhibition, in A.J. Ijspeert, M. Murata, N. Wakamiya (Eds.), Biologically Inspired Approaches to Advanced Information Technology: First International Workshop, BioADIT 2004, Lausanne, Switzerland, 2004, Revised Selected Papers, Lecture Notes in Computer Science, 2004, vol. 3141, pp. 166-175.
Y. Yu, Y. Choe, Angular disinhibition effect in a modified Poggendorff illusion, in K.D. Forbus, D. Gentner, T. Regier (Eds.), Proceedings of the 26th Annual Conference of the Cognitive Science Society, Mahwah, NJ, USA, (2004), pp. 1500-1505.
M. Wax, T. Kailath, Efficient inversion of Toeplitz-block Toeplitz matrix, IEEE Trans. Acoust, Speech Signal Process., (31) 5 (1983), pp. 1218-1221.
N. Kalouptsidis, G. Carayannis, and D. Manolakis, On block matrices with elements of special structure, Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP’82, Paris, France, (May 1982), vol. 7, pp. 1744-1747.
A.E. Yagle, A fast algorithm for Toeplitz-block-Toeplitz linear systems, Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, ICASSP’01, Salt Lake City, USA, (2001), vol. 3, pp. 1929-1932.
S.J. Reeves, Fast algorithm for solving block banded Toeplitz systems with banded Toeplitz blocks, Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, ICASSP’02, Orlando, USA, (2002), vol. 4, pp. 3325-3328.
D.A. Bini, B. Meini, Solving block banded block Toeplitz systems with banded Toeplitz blocks, in F.T. Luk (Ed.), Proc. SPIE, Advanced Signal Processing Algorithms, Architectures, and Implementations IX, 1999, vol. 3807 pp. 300-311.
U. Grenander, G. Szegö, Toeplitz Forms and Their Applications, 2nd ed., Chelsea Publishing, New York, 1984.
S. Serra, Preconditioning strategies for asymptotically ill conditioned block Toeplitz systems, BIT Num. Math., (34) (1994), pp. 579-594.
P. Tilli, On the asymptotic spectrum of Hermitian block Toeplitz matrices with Toeplitz blocks, Math. Comp., (66) 219 (1997), pp. 1147-1159.
W.T. Freeman, H.H. Adelson, The design and use of steerable filters, IEEE Trans. Pattern Anal. Mach. Intell., (13) 9 (1991), pp. 891-906.
S. Barnet, Matrices: Methods and Applications, Oxford University Press, New York, 1990.
R.A. Young, L.M. Lesperance, W.W. Meyer, The Gaussian derivative model for spatial-temporal vision: I. Cortical model, Spat. Vis., (14) 3,4 (2001), pp. 261-319.
E.M. Stein, G. Weiss, Introduction to Fourier Analysis on Euclidean Spaces, Princeton University Press, Princeton, New Jersey, 1971.
J.B. Price, M.H. Hayes, Steerable filter cascades, Proc. IEEE Intentional Conference on Image Processing ICIP’99, Kobe, Japan, (October 1999), vol. 2, pp. 880-884.
Downloads
Published
How to Cite
Issue
Section
License
International Journal of Computing is an open access journal. Authors who publish with this journal agree to the following terms:• Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
• Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
• Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
