Image analysis – Learning systems – Neural networks
Patent
1995-05-22
1997-04-29
Johns, Andrew
Image analysis
Learning systems
Neural networks
382216, 395 23, G06T 140
Patent
active
056257074
ABSTRACT:
Pattern recognition, for instance optical character recognition, is achieved by training a neural network, scanning an image, segmenting the image to detect a pattern, preprocessing the detected pattern, and applying the preprocessed detected pattern to the trained neural network. The preprocessing includes determining a centroid of the pattern and centrally positioning the centroid in a frame containing the pattern. The training of the neural network includes randomly displacing template patterns within frames before applying the template patterns to the neural network.
REFERENCES:
patent: 3760356 (1973-09-01), Srivastava
patent: 3846752 (1974-11-01), Nakano et al.
patent: 3930231 (1975-12-01), Henrichon, Jr. et al.
patent: 3993976 (1976-11-01), Ginsburg
patent: 4225850 (1980-09-01), Chang et al.
patent: 4241329 (1980-12-01), Bahler et al.
patent: 4346405 (1982-08-01), Yoda et al.
patent: 4513441 (1985-04-01), Henshaw
patent: 4547800 (1985-10-01), Masaki
patent: 4817176 (1989-03-01), Marshall et al.
patent: 4958939 (1990-09-01), Samad
patent: 5003490 (1991-03-01), Castelaz et al.
patent: 5121443 (1992-06-01), Tomlinson
patent: 5161203 (1992-11-01), Buckley
patent: 5245672 (1993-09-01), Wilson et al.
patent: 5263097 (1993-11-01), Katz et al.
patent: 5287272 (1994-02-01), Rutenberg et al.
patent: 5305392 (1994-04-01), Longest, Jr. et al.
patent: 5323471 (1994-06-01), Hayashi
patent: 5452402 (1995-09-01), Sakiyama et al.
patent: 5475768 (1995-12-01), Diep et al.
Gonzalez et al., Digital Image Processing, Addison-Wesley Publishing (1992), pp. 443-445 and 595-616 1992.
Elliman, et al., "Shift Invariant Neural Net for Machine Vision," IEEE Proceedings, vol. 137, Pt. I, No. 3, pp. 183-187 Jun. 1990.
Gurgen et al., "On the Training Strategies of Neural Networks for Speech Recognition," Speech and Acoustics Lab, Tokyo, Japan (none).
A. Khotanzad, et al., "Distortion Invariant Character Recognition by a Multi-Layer Perceptron and Back-Propagation Learning," IEEE International Conference on Neural Networks, Jul. 24-27, 1988, San Diego, California, U.S.A, pp. 625-632.
A. Khotanzad and J. Lu, "Classification of Invariant Image Representations Using a Neural Network," 38 IEEE Transactions on Acoustics, Speech and Signal Processing 6, Jun. 1990,U.S.A., pp. 1028-1038.
Gerhard Seiler, "Small Object Counting With Cellular Neural Networks," 1990 IEEE International Workshop on Cellular Neural Networks and Their Applications CNNA-90, Dec. 16-19, 1990, Budapest, Hungary, pp. 115-123.
S.N. Srihari, et al., "Pattern Recognition, Character Recognition, and Optical Character Readers," Technical Report CEDAR-TR-91-1; Center for Document Analysis and Recognition, May 1991, State University of New York at Buffalo, N.Y., pp. 1-51.
Katsuji Imai, et al, "Pattern Extraction and Recognition for Noisy Images Using the Three-Layered BP Model," 1991 IEEE International Joint Conference on Neural Networks, vol. 1, Nov. 18-21, 1991, Singapore, pp. 262-267.
J. Loncelle, et al., "Optical Character Recognition an Cooperating Neural Networks Techniques," Artificial Neural Networks, 2, I. Aleksander and J. Taylor, Ed., Elsevier Science Publishers, B.V., Netherlands, 1992, pp. 1591-1595.
Glenn S. Himes, et al., "Centroid calculation using neural networks," Journal of Electornic Imaging, vol. 1, Jan. 1992, U.S.A., pp. 73-87.
Shingchern D. You, et al., "Connectionist model for object recognition," Applications of Artificial Neural Networks III, SPIE, vol. 1709, Apr. 21-24, 1992, Orlando, Florida, U.S.A., pp. 200-207.
Ofer Matan, et al., "Reading Handwritten Digits: A Zip Code Recognition System", Computer, Jul. 1992, vol. 25, No. 7, IEEE Computer Society, U.S.A., pp. 59-63.
Avi-Itzhak Hadar I.
Diep Thanh A.
Garland Harry T.
Canon Inc.
Johns Andrew
Meyer Stuart P.
Radlo Edward J.
LandOfFree
Training a neural network using centroid dithering by randomly d does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Training a neural network using centroid dithering by randomly d, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Training a neural network using centroid dithering by randomly d will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-712647