Title :
Application of the ANNA neural network chip to high-speed character recognition
Author :
Säckinger, Eduard ; Boser, Bernhard E. ; Bromley, Jane ; LeCun, Yann ; Jackel, Lawrence D.
Author_Institution :
AT&T Bell Labs., Holmdel, NJ, USA
fDate :
5/1/1992 12:00:00 AM
Abstract :
A neural network with 136000 connections for recognition of handwritten digits has been implemented using a mixed analog/digital neural network chip. The neural network chip is capable of processing 1000 characters/s. The recognition system has essentially the same rate (5%) as a simulation of the network with 32-b floating-point precision
Keywords :
application specific integrated circuits; character recognition; computerised pattern recognition; neural nets; ANNA neural network chip; floating-point precision; handwritten digits; high-speed character recognition; mixed analog/digital neural network; recognition system; Arithmetic; Character recognition; Handwriting recognition; High speed optical techniques; Logic testing; Neural networks; Neurons; Optical character recognition software; Optical computing; Optical network units;
Journal_Title :
Neural Networks, IEEE Transactions on
