NNORC-a neural network for the optical character recognition

Neural networks (NN) provide a promising approach to solve problems which are very hard for classical artificial intelligence, in fact due to their generalization capability and noise insensitivity, they have been applied to pattern recognition with very encouraging results. In this paper a multilayer backpropagation neural network for optical character recognition (NNORC) is presented. NNORC acquires data from a scanner system, recognizes the typewritten characters, assigns them an ASCII code and stores the informations in a computer file, accessible from any available program for further processing such as editing databases and cross-checking comparisons. The implemented system NNORC runs on PC 486-33 MHZ and starting from documents of omnifonts type of the Latin alphabet recognises 20 characters per second