Abstract
This paper describes an architectural extension of the shared-weight neural network (SWNN) that performs shift-invariant filtering using fuzzy-morphological operations for feature extraction. The nodes in this stage employ the generalized-mean operator to implement fuzzymorphological operations. The network parameters, weights, morphological structuring element and fuzziness, are optimized through error back-propagation(EBP) training method.
The parameter values of the trained SWNN are then implanted into the extended SWNN (ESWNN) which is a simple correlational type neural network. This architecture dramatically reduces the amount of computation by avoiding segmentation process.
The neural network is applied to automatic recognition of a vehicle in visible images of a parking lot. The network is tested with several sequences of images that include targets ranging from no occlusion to almost full occlusion. The results demonstrate an ability to detect occluded targets, while trained with non-occluded ones. In comparison, the proposed network produces better results than the ordinary SWNN.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Y. le Cun, L. D. Jackel, B. Boser, J. S. Denker, H. P. Graf, I. Guyon, D. Henderson, R. E. Howard, and W. Hubbard, “Handwritten Digit Recognition: Application of Neural Network Chips and Automatic Learning.” IEEE Communications Magazine, November, pp. 41–64, 1989
H. Dyckhoff and W. Pedrycz, “Generalized Means as a Model of Compensation Connectives.” Fuzzy Sets and Systems, Vol. 14, No. 2, pp. 143–154, 1984.
D.E. Rumelhart, G.E. Hinton and R.J. Williams, “Learning Internal Representations by Error Propagation.” in D.E. Rumelhart and J.L. McClelland, editors, Parallel Distributed Processing:Explorations in the Microstructure of Cognition, MIT Press, Cambridge, MA, 1986.
E.R. Dougherty, An Introduction to Morphological Image Processing, SPIE Optical Engineering Press, 1992.
J. Serra, Image Analysis and Mathematical morphology, Vol. 1, Academic Press, New York, 1982.
Y. Won and B-H Lee, “A Fuzzy Morphological Neural Network: Principles and Implementation,” Trans. Korea Information Processing Society, Vol. 3, No. 3, pp. 449–459, 1995
P.D. Gader, Y. Won and M.A. Khabou, “Image Algebra Networks for Pattern Classification,” SPIE Mathematical Morphology and Image Algebra, pp. 157–168, July, 1993.
P. Gader, J. Miramonti, Y. Won and P. Coffield, “Segmentation-free Neural Network for Automatic Vehicle Detection”, Neural Networks, Vol. 8, No 9 pp 1457–1473, 1995
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Won, Y., Lee, BH., Baek, YC., Lee, JS. (1997). Shift-invariant fuzzy-morphology neural network for occluded target recognition. In: Mira, J., Moreno-Díaz, R., Cabestany, J. (eds) Biological and Artificial Computation: From Neuroscience to Technology. IWANN 1997. Lecture Notes in Computer Science, vol 1240. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0032574
Download citation
DOI: https://doi.org/10.1007/BFb0032574
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63047-0
Online ISBN: 978-3-540-69074-0
eBook Packages: Springer Book Archive