Neural network with back propagation controlled through an outpu

Details Neural network with back propagation controlled through an outpu Neural network with back propagation controlled through an outpu

1990-05-07

1991-09-24

Boudreau, Leo H.

Image analysis

Histogram processing

For setting a threshold

382 27, 3642749, 364513, G06K 966

Patent

active

050520431

ABSTRACT:
Apparatus, and an accompanying method, for a neural network, particularly one suited for use in optical character recognition (OCR) systems, which through controlling back propagation and adjustment of neural weight and bias values through an output confidence measure, smoothly, rapidly and accurately adapts its response to actual changing input data (characters). Specifically, the results of appropriate actual unknown input characters, which have been recognized with an output confidence measure that lies within a pre-defined range, are used to adaptively re-train the network during pattern recognition. By limiting the maximum value of the output confidence measure at which this re-training will occur, the network re-trains itself only when the input characters have changed by a sufficient margin from initial training data such that this re-training is likely to produce a subsequent noticeable increase in the recognition accuracy provided by the network. Output confidence is measured as a ratio between the highest and next highest values produced by output neurons in the network. By broadening the entire base of training data to include actual dynamically changing input characters, the inventive neural network provides more robust performance than which heretofore occurs in neural networks known in the art.

REFERENCES:
patent: 3192505 (1965-06-01), Rosenblatt
patent: 3267439 (1966-08-01), Bonner
patent: 3275985 (1966-09-01), Dunn et al.
patent: 3275986 (1966-09-01), Dunn et al.
patent: 3646329 (1972-02-01), Yoshino et al.
patent: 4479241 (1984-10-01), Buckley
patent: 4504970 (1985-03-01), Werth et al.
patent: 4682365 (1987-07-01), Orita et al.
patent: 4742556 (1988-05-01), Davis, Jr. et al.
patent: 4748674 (1988-05-01), Freeman
patent: 4873661 (1989-10-01), Tsividis
patent: 4876731 (1989-10-01), Loris et al.
patent: 4884216 (1989-11-01), Kuperstein
patent: 4885757 (1989-12-01), Provence
patent: 4912649 (1990-03-01), Wood
patent: 4912651 (1990-03-01), Wood et al.
patent: 4912652 (1990-03-01), Wood
patent: 4912654 (1990-03-01), Wood
patent: 4912655 (1990-03-01), Wood
patent: 4918618 (1990-04-01), Tomlinson, Jr.
patent: 4921647 (1990-03-01), Wood
patent: 4933872 (1990-01-01), Vandenberg et al.
patent: 4951239 (1990-11-01), Andes et al.
Philip D. Wasserman, "Neural Computing--Theory & Practice", 1989, pp. 128-129.
N. J. Nilsson, The Mathematical Foundations of Learning Machines (.COPYRGT. 1990: Morgan Kaufmann Publishers; San Mateo, Calif.) and particularly section 2.6 "The Threshold Logic Unit (TLU)", on pp. 21-23 and Chapter 6 Layered Machines on pp. 95-114.
G. L. Martin et al., "Recognizing Hand-Printed Letters and Digits Using Backpropagation Learning", Technical Report of the MCC, Human Interface Laboratory, Austin, Tex., Jan. 1990, pp. 1-9.
J. S. N. Jean et al., "Input Representation and Output Voting Considerations for Handwritten Numeral Recognition with Backpropagation", International Joint Conference on Neural Networks, Washington, D.C., Jan. 1990, pp. I-408 to I-411.
X. Zhu et al., "Feature Detector and Application to Handwritten Character Recognition", International Joint Conference on Neural Networks, Washington, D.C., Jan. 1990, pp. II-457 to II-460.
K. Haruki et al., "Pattern Recognition of Handwritten Phonetic Japanese Alphabet Characters", International Joint Conference on Neural Networks, Washington, D.C., Jan. 1990, pp. II-515 to II-518.
R. K. Miller, Neural Networks (.COPYRGT. 1989: Fairmont Press; Lilburn, Ga.), pp. 2-12 and Chapter 4 "Implementation of Neural Networks", on pp. 4-1 to 4-26.
Y. Hayashi et al., "Alphanumeric Character Recognition Using a Connectionist Model with the Pocket Algorithm", Proceedings of the International Joint Conference on Neural Networks, Washington, D.C. Jun. 18-22, 1989, vol. 2, pp. 606-613.
M. Caudill, "Neural Networks Primer--Part III", AI Expert, Jun. 1988, pp. 53-59.
D. J. Burr, "A Neural Network Digit Recognizer", Proceedings of the 1986 IEEE International Conference of Systems, Man and Cybernetics, Atlanta, Ga., pp. 1621-1625.
D. E. Rumelhart, et al., Parallel Distributed Processing, (.COPYRGT. 1986: MIT Press; Cambridge, Mass.), and specifically Chapter 8 thereof "Learning Internal Representations by Error Propagation", pp. 318-362.

Affiliated with

Also associated with

Rating

Neural network with back propagation controlled through an outpu does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Neural network with back propagation controlled through an outpu, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neural network with back propagation controlled through an outpu will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1703736