Un Syst Me Stochastique De Reconnaissance De Caract Res Imprim S Multipolices A Stochastic Recognition System Of Printed Characters Of Any Font And Size

Un système stochastique de reconnaissance de caractères imprimés multipolices [A stochastic recognition system of printed characters of any font and size]

A Fast Recognition System for Isolated Printed Characters

The purpose of this book is to propose a new multi stage algorithm for the recognition of isolated characters. The algorithm can be applied for any isolated character such as Latin, Chinese, Japanese, and Arabic characters but it has been applied in this paper for Arabic characters, since the Arabic characters OCR systems are rarely considered in literature, so the challenge was to use the system on non-Latin script .The approach uses normalized and isolated characters of the same size and extracts an image signature based on the center of gravity of the character and then compares this values to a set of signatures for typical characters of the set. The system then provides the closeness of match to all other characters in the set which is expressed in a triangular confusion matrix. This algorithm is believed to be new and, to the contrary of earlier investigations, its accuracy and running time can be tuned to get the best strategy for any character set.

An Adaptive Recognition System for Machine Printed Characters Using Microcomputers