Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!dynas!neural!egel From: egel@neural.dynas.se (Peter Egelberg) Newsgroups: comp.ai.neural-nets Subject: Re: Neural Networks To Recognize Handwritten Characters Message-ID: <1991Feb27.084553.4019@neural.dynas.se> Date: 27 Feb 91 08:45:53 GMT References: <88777@tut.cis.ohio-state.edu> Organization: /etc/organization Lines: 25 In article <88777@tut.cis.ohio-state.edu> writes: >Hi. I'm new to neural networks, and I'd appreciate some references on >using a multi-layer NN with the bacpropagation algorithm to recognize >hand written charcaters. > >Thanks in advance. > >Saad Mufti (mufti@cis.ohio-state.edu) LeCun Y. and others 1989 "Backpropagation Applied to Handwritten Zip Code Recognition". In Neural Computation 1, 541-551. Fukushima K. and others 1983 "Neocognitron: a neural network model for a mechanism of visual pattern recognition". In IEEE Transactions on Systems, Man and Cybernetics SMC-13:826-834 Fukushima doesn't use backpropagation but the results are impressive. Unfortunately the size of the network makes it impractical for applications but it can give you some ideas. -- Peter Egelberg E-mail: egel@neural.dynas.se Neural AB Phone: +46 46 11 00 90 Otto Lindbladsv. 5 Fax: +46 46 13 60 85 223 65 LUND, SWEDEN