Neural network methods show great promise for providing highly accurate, noise resistant, parallel algorithms and data organization for a wide range of problems where “humanlike” recognition ability is needed. One specific area of recognition, the conversion of images of handwritten to computer representation, is used in this paper to explore the two major types of machine learning—supervised learning and self-organization—and to demonstrate the capabilities of neural network algorithms.
Get full access to this article
View all access options for this article.
References
1.
AndersonJ. A., & RosenfeldE. (1989). Neurocomputing.Cambridge, MA: MIT Press.
2.
AndersonJ. A., RossenM. L., ViscusoS. R., & SerenoM. E. (1990). Experiments with the representation in neural networks: Object, motion, speech, and arithmetic. In FankenH., and StadlerM. (eds.), Synergetics of cognition (54–69). Berlin: Springer-Verlag.
3.
BeauchampK. G. (1975). Walsh functions and their applications.London: Academic Press.
4.
CarpenterG. A., & GrossbergS. (1987). A massively parallel architecture for a self-organizing neural pattern recognition machine.Computer Vision, Graphics, and Image Processing37: 54–115.
5.
DaugmanJ. G. (1988). Complete discrete 2-D Gabor transform by neural networks for image analysis and compression.IEEE Trans, on Acoustics, Speech, and Signal Processing36: 1169–79.
6.
DaugmanJ. G. (1985). Representational issues and local filter models of 2-D spatial visual encoding. In RoseD., and DobsonV. G. (eds.), Models of the visual cortex (96–107). New York: J. Wiley and Sons.
7.
GaborD. (1946). Theory of communication.Journal of the Institute of Electrical Engineers, 3 (93): 429–57
8.
GrossbergS. (1969). On learning and energy-entropy dependence in recurrent and nonrecurrent signed networks.J. Statistical Physics1: 319–50.
9.
GrotherP. Forthcoming. Karhunen Loève feature extraction for neural handwritten character recognition.
10.
JackelL. D., GrafH. P., HubbardW., DenkerJ. S., HendersonD., & GuyonI. (1988). An application of neural net chips: Handwritten digit recognition. In IEEE International Conference on Neural Networks, vol. 2 (107–15). San Diego.
11.
KohonenT. (1988). Self-organization and associative memory.2d ed.Berlin: Springer-Verlag.
12.
LachenbruchP. A. (1982). Discriminant analysis. In KotzS., JohnsonN. L., & ReadC. B. (eds.), Encyclopedia of Statistical Sciences (389–97). New York: Wiley-Inter-science.
13.
LinskerR. (1988). Self-organization in a perceptual network.Computer21: 105–17.
14.
LippmanR. P. (1987). An introduction to computing with neural nets.IEEE ASSP Magazine, April 4–22.
15.
McCullochW. S., & PittsW. (1943). A logical calculus of the ideas immanent in nervous activity.Bull. Math. Biophysics9: 115–33.
16.
MinskyM., & PapertS. (1969). Perceptrons.Cambridge: MIT Press.
17.
RajaveluA., MusaviM. T., & ShirvaikarM. V. (1989). A neural network approach to character recognition.Neural Networks2: 387–93.
18.
RojerA., & SchwatzE. (1989). Multi-map model for pattern classification.Neural Computation1: 104–15.
19.
RosenblatF. (1958). The perceptron: A probabilistic model for information storage and organization in the brain.Psychological Review65: 386–408.
20.
RubnerJ., & SchultenK. (1990). Development of feature detectors by self-organization.Biological Cybernetics62: 193–99.
21.
RumelhartD. E., HintonG. E., & WilliamsR. J. (1988). Learning internal representations by error propagation. In RumelhartD. E., and McClellandJ. L., etal. (eds.) Parallel distributed processing: Explorations in the microstructure of cognition. Vol. 1, Foundations, ch. 8 (318–62). Cambridge: MIT Press.
22.
SpechtD. F. (1990). Probabilistic neural networks.Neural Networks, 3 (1): 109–18.
23.
WilsonC. L. Forthcoming. A new self-organizing neural network architecture for parallel multi-map pattern recognition—FAUST.Progress in Neural Networks, 4.
24.
WilsonC. L., WilkinsonR. A., & GarrisM. D.(1990, June). Self-organizing neural network character recognition on a massively parallel computer. In Proc. of the IJCNN, vol. 2 (325–29).
25.
WinstonP. H. (1984). Artificial Intelligence.Reading, MA: Addison-Wesley.
